Gene deletion in barley mediated by LTR-retrotransposon BARE

1 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 www.na u e.com/scien i ic epo s Gene Dele ion in Ba ley Media ed by LTR- e o ansposon BARE Yi Shang1,*, Fei Yang2,3,*, Alan H. Schulman4,5, Jinghuan Zhu1, Yong Jia3, Junmei Wang1, Xiao-Qi Zhang3, Qiaojun Jia6, Wei Hua1, Jianming Yang1 & Chengdao Li2,3 A poly- ow b anched spike (p bs) ba ley mu an was ob ained om soaking a wo- owed ba ley in lo escence in a solu ion o maize genomic DNA. Posi ional cloning and sequencing demons a ed ha he p bs mu an esul ed om a 28 kb dele ion including he in lo escence a chi ec u e gene H RA2. Sequence anno a ion e ealed ha he H RA2 gene is lanked by wo LTR (long e minal epea ) e o ansposons (BARE) sha ing 89% sequence iden i y. A ecombina ion be ween he in eg ase (IN) gene egions o he wo BARE copies esul ed in he o ma ion o an in ac BARE and loss o H RA2. No maize DNA was de ec ed in he ecombina ion egion al hough he lanking sequences o H RA2 gene showed o e 73% o sequence iden i y wi h epe i i e sequences on 10 maize ch omosomes. I is s ill unknown whe he he in e ac ion o e o ansposons be ween ba ley and maize has esul ed in he ecombina ion obse ed in he p esen s udy. The a chi ec u e o b anched in lo escences in g asses depends on he de elopmen al a e o p imo dia and axis o ien a ion1. The ice (O yza sa i a L.) panicle gene a es se e al p ima y and seconda y b anches on which spike- le s a e p oduced. So ghum and maize male in lo escences sha e a s uc u e simila o ha o ice. In ba ley (Ho deum ulga e L.) spikes, howe e , spikele s a e bo ne di ec ly on he main axis, he achis, and he e a e no pedicels. A diagnos ic ea u e o ba ley is he possession o h ee one- lowe ed spikele s a each achis node2,3. Based on la e al spikele size and e ili y, ba ley is classi ied in o wo- owed and six- owed ypes. Two- owed ba ley only has a cen al e ile spikele wi h small and in e ile la e al spikele s while he six- owed ba ley has h ee ully-de eloped e ile spikele s. The majo genes ha con ol ow- ype a ia ion in ba ley a e V s14, In -c5 and V s46. The ba ley domes ica- ion gene V s1, loca ed on he long a m o ch omosome 2H, encodes a homeodomain-leucine zippe (HD-Zip) ansc ip ion ac o ha supp esses he de elopmen o la e al spikele s in wo- owed ba ley. Mu an s1 esul s in a well-de eloped six- owed pheno ype4. In -c, loca ed on ch omosome 4H, is an o holog o he maize (Zea mays. L.) domes ica ion gene, Teosin e b anched 1 (TB1), a membe o he TCP gene amily encoding pu a i e basic helix-loop-helix DNA-binding p o eins5. The In -c gene modi ies la e al spikele e ili y in ba ley, and can in luence he pheno ypic e ec o he V s1 locus7. V s4 con ols ow- ype and spikele de e minacy in ba ley; an induced mu a ion, s4, can con e he wo- owed o a six- owed pheno ype5,8. V s4 is an o holog o he maize in lo escence a chi ec u e gene RAMOSA2 (RA2), which encodes a ansc ip ional egula o ha con ains he la e al o gan bounda ies (LOB) domain. Exp ession analyses by mRNA in si u hyb idiza ion and mic oa ay app oaches showed ha V s4 is exp essed e y ea ly du ing in lo escence de elopmen and con ols he ow- ype pa hway h ough V s1 by nega i ely egula ing he la e al spikele e ili y in ba ley. Mo eo e , he V s4 gene is an impo an modi ie o in lo escence de elopmen . He e, we epo on a new mu an , poly- ow and b anched spike (p bs) ob ained by soaking a wo- owed ba ley in lo escence in maize genomic DNA om a single c oss hyb id9,10, and cha ac e ize i s gene ics, epo i s posi ional cloning, and analyze i s o igin. 1Na ional Ba ley Imp o emen Cen e, Ins i u e o C op and Nuclea Technology U iliza ion, Zhejiang Academy o Ag icul u al Science, Hangzhou 310021, China. 2Depa men o Gene ics and Cell Biology, Yang ze Uni e si y, Jingzhou, Hubei 434023, China. 3Wes e n Ba ley Gene ics Alliance, Mu doch Uni e si y, 90 Sou h S ee , Mu doch WA 6150, Aus alia. 4Luke/BI Plan Genomics Lab, Ins i u e o Bio echnology and Viikki Plan Science Cen e, Uni e si y o Helsinki, FIN-00014 Helsinki, Finland. 5G een Technology, Na u al Resou ces Ins i u e Finland (Luke), Viikinkaa i 1, FIN-00790 Helsinki, Finland. 6College o Li e Sciences, Zhejiang Sci-Tech Uni e si y, Hangzhou 310018, China. *These au ho s con ibu ed equally o his wo k. Co espondence and eques s o ma e ials should be add essed o J.Y. (email: [email p o ec ed]) o C.L. (email: [email p o ec ed]) Recei ed: 23 Augus 2016 Accep ed: 27 Janua y 2017 Published: 02 Ma ch 2017 OPEN www.na u e.com/scien i ic epo s/ 2 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 Resul s Mu an p bs esul ed om dele ion o he V s4 gene. The poly- ow and b anched spike (p bs) ba - ley mu an was ob ained by soaking a wo- owed ba ley in lo escence in maize genomic DNA solu ion9,10. The mu an p bs no only changes wo- owed ba ley in o a poly- owed o m bu also adds a spikele ow, o ming i egula poly- ow and b anched spikes (Fig.1). Gene ic analysis indica ed ha he mu an pheno ype was caused by a ecessi e gene, which has an epis a ic e ec on V s111. The p bs was ini ially mapped o he cen ome e o he sho a m o ch omosome 3H11,12, a loca ion simila o ha o s4. Fu he mo e, he imma u e spikes o he p bs mu an unde s e eoscope a e akin o he scanning elec on mic oscopy images o he V s4 imma u e spikes12. Th ee molecula ma ke s (DQ327702, Cbic43, and Cbic44), closely linked wi h he V s4 gene, co-seg ega ed wi h he p bs gene in he p bs/Kunlun12 RIL and p bs/Zangqing 320 F2 popula ions (Supplemen a yFig.1). No agmen was ampli ied in he mu an plan s using hese h ee molecula ma ke s. Simila ly, h ee o he p ime s co e ing he V s4 gene sequence also ailed o ampli y a speci ic DNA agmen om ei he he p bs mu an o i s p ogeny 11R258-95. Exp ession analyses e ealed ha he exp ession o V s4 was no de ec ed and he exp es- sion o V s1 was signi ican ly down- egula ed in imma u e spikes a lemma p imo dium s age o he p bs mu an (Fig.2). These esul s indica ed ha he p bs mu an may ha e esul ed om a la ge dele ion a ound he V s4 gene. Iden i ica ion o dele ion egion in p bs mu an . To iden i y he dele ion egion in he p bs mu an , a Mo ex BAC clone was iden i ied ha con ains he V s4 gene. PCR p ime s we e designed a 2 kb in e als om 14 kb ups eam o 22 kb downs eam o he V s4 gene and we e es ed on he p bs mu an , 11R258-95, Pudamai-2, Figu e 1. Mo phology o de eloping and ma u e spikes. (a) no mal spikes; (b,c and d) b anching spikes; (e) b anches o (d, ) mu an p bs; (g) wild pa en Pudamai-2. Figu e 2. Rela i e exp ession le el o V s4 and V s1 de e mined by quan i a i e RT-PCR in imma u e spikes a lemma p imo dium s age o he p bs mu an and wild pa en Pudamai-2. Cons i u i ely exp essed H Ac in was used o no maliza ion. www.na u e.com/scien i ic epo s/ 3 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 and Mo ex. PCR p ime s loca ed in he egion om 3 kb ups eam o 10 kb downs eam o he V s4 gene ailed o ampli y a speci ic DNA agmen in he p bs mu an and 11R258-95 bu ampli ied a single band in Pudamai-2 and Mo ex ins ead. Sequencing e ealed ha amplicons ep esen ed a single p oduc in Pudamai-2 and Mo ex. P ime s designed om 3 o 13 kb ups eam and 10 o 21 kb downs eam o he V s4 gene ampli ied a single band in all es ed plan s, bu he amplicons ep esen ed mul iple p oduc s when sequenced. These esul s did no sup- po hese egions a ising om a single dele ion e en in he p bs mu an . Howe e , an addi ional p ime pai , Cbic123, ma ching a si e 14 kb ups eam o V s4, ampli ied a single band in all es ed plan s; he PCR p oduc had 100% sequence iden i y among he p bs mu an , 11R258-95, Pudamai-2, and Mo ex. Ano he p ime pai Cbic119, 22 kb downs eam o V s4, also ampli ied a single band in all es ed plan s (sequencing was iden ical in all es ed lines). These esul s e ealed ha he dele ion sequence was om ~13 o 36 kb in he p bs mu an . A e ailu e o ampli y a single DNA agmen using many PCR p ime s in he a ge egion, long- ange PCR was used o isola e he sequence co e ing he p bs mu a ion. Based on he abo e PCR es esul s, PCR p ime s Cbic131 and Cbic132 we e designed o his pu pose; he o wa d p ime was nea he si e o p ime Cbic123 and he e e se p ime nea he si e o p ime Cbic119, as bo h ha e been con i med o ampli y a single copy o DNA om he con ol a ie ies and mu an s. A 15 kb agmen was success ully ampli ied om bo h he p bs mu an and 11R258-95 (Fig.3), whe eas he con ol PCRs using DNA om Pudamai-2 and Mo ex as a empla e ailed o ampli y. The ampli ica ion p oduc om p bs was 14,715 bp (accession numbe KU758926). We iden i ied a 48,951 bp sequence om Mo ex using he 14,715 bp p bs sequence as que y in BLASTN sea ches o he Mo ex genome da abase (h p://webblas .ipk-ga e sleben.de/ba ley/ i oblas .php). Alignmen o hese wo sequences showed ha a 27,804 bp sequence in Mo ex, ex ending om n 15,680 o n 43,769 and con aining he en i e V s4 gene, is dele ed in p bs (Fig.4). Sequence analysis demons a ed ha he V s4 gene in Mo ex is lanked by wo long e minal epea (LTR) e o ansposons loca ed, espec i ely, a n 10,967 o 19,828 ups eam and n 38,791 o n 47,684 downs eam (Fig.4). A sea ch o he T i iceae Repe i i e Elemen s (TREP) Figu e 3. PCR-ampli ica ion o he 15-kb ecombina ion sequence agmen in he p bs mu an and RIL 11R258-95. www.na u e.com/scien i ic epo s/ 4 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 da abase e ealed high simila i y o he e o ansposon RLC BARE1 B consensus-1 (TREPACC = TREP3133) wi h 90% and 98% o sequence iden i y, espec i ely, o he wo elemen s. We e e o he e o ansposon ups eam o V s4 as BARE up, and he one downs eam o V s4 as BARE down. The wo sha e 90% iden i y, a e bound by 1.8 kb LTRs, and con ain he ull-leng h open eading ame encoding Gag, aspa ic p o einase (AP), in eg ase (IN), e e se ansc ip ase (RT), and RNaseH (RH) expec ed o canonical BARE1 elemen s13 (Fig.4). We u he sequenced pa o he dele ion om he wild pa en Pudamai-2, which con ains a comple e V s4 gene. A phylogene ic ee was cons uc ed using he MEGA6 p og am14 wi h he minimum e olu ion me hod. The esul s showed ha he V s4 haplo ype in Pudamai-2 was simila o haplo ype 8 (Supplemen a yFig.2) which was cha ac e ized by an inse ion o TA bases in he 5′ UTR o he gene6 and was mainly dis ibu ed in Asia. Thus, he p bs mu an esul ed om a dele ion o he en i e unc ional gene H RA2, a ba ley o holog o he maize in lo escence a chi ec u e gene RAMOSA2 (RA2), which he eby ans o med a wo- ow ba ley in o a poly- ow b anched s uc u e. Recombina ion be ween he in eg ase genes o wo BAREs o med p bs. Sequence analysis o he 15 kb egion om p bs iden i ied a single BARE elemen o he canonical 8.9 kb in leng h. Alignmen e ealed ha he 5′ pa o he p bs (n 1 o 4,714) BARE was iden ical o BARE up, whe eas he 3′ pa (4,972 o 8,918 bp) was he same as BARE down (Fig.5). The join be ween he wo hal es is be ween n 4,715 and n 4,971 in he Figu e 4. Schema ic compa ison he Mo ex and p bs mu an sequences. The V s4 gene is lanked by wo BARE e o ansposons which a e bounded by 1.8 kb long e minal epea s (LTRs) and con ains coding domains o GAG, in eg ase (IN), e e se ansc ip ase (RT), and RNaseH in Mo ex. The wo BAREs sequences a e highly conse ed and sha e 89% o sequence iden i y. The p bs mu an has 28 kb dele ion including he en i e V s4 gene and pa s o he wo e o ansposons. The non- e o ansposon sequence in he wild pa en Pudamai 2 sha e 100% o sequence iden i y wi h Mo ex. Figu e 5. Alignmen o he h ee LTR e o ansposons: BARE up, BARE down, BARE 3, shows ecombina ion egion. Red capi al le e s ma ked Indel si es and black lowe le e s ma ked consensus sequences. Yellow backg ound show BARE up sequences and hei homologous sequences in BARE 3. G een backg ound show BARE down sequences and hei homologous sequences in BARE 3. Pu ple ed show he ecombina ion egion decided by he alignmen . www.na u e.com/scien i ic epo s/ 5 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 p bs BARE (Fig.5), wi hin he in eg ase (IN) domain, co esponding o posi ion n 9,402–9,690 in he cloned p bs agmen (accession numbe KU758926). Re o ansposon in eg a ion gene a es a di ec - epea a ge -si e duplica ion (TSD) lanking he indi idual elemen , as a consequence o epai o he s agge ed cu made by he in eg ase15. BARE up is lanked by impe ec CCAAG TSDs and BARE down by a pe ec pai o CTGAA mo i s. The BARE in p bs is lanked by CCAAG and CTGAA, suppo ing he o igin o his BARE by ecombina ion. Mo eo e , he ups eam and downs eam sequences su ounding he single BARE in p bs co espond, espec- i ely, o ha ups eam o BARE up and downs eam o BARE down. Thus, he H RA2 gene, lanked by wo BARE elemen s in Pudamai-2, was dele ed by ecombina ion be ween hem, he eby gene a ing he p bs mu a ion. Sea ch he maize genome o sequence simila i y. To in es iga e he possible ole o maize DNA in o ming he p bs mu a ion, we used he egion spanning he ecombina ion zone in he BARE elemen s (9,402– 9,690 bp) o ca y ou a BLASTn sea ch he MaizeGDB B73 e e ence genome sequence. No maize-speci ic sequence was ound in he egion. Hence, i appea s ha no maize DNA has been inse ed in o he p bs egion and ha he p bs pheno ype esul s solely om he V s4 gene dele ion. As an al e na i e o inse ion, he maize DNA may ha e played a ole h ough sequence simila i y a he ecombina ion poin . The egion o ecombina ion in p bs co esponds o he mos conse ed pa o he in e- g ase gene, which is he co e domain ha includes he D-D-35-E ac i e si e mo i 16. The ecombina ion i sel ook place in he egion be ween he second Asp and he Glu o he ac i e si e; a BLASTn sea ch o his egion agains he MaizeGDB B73 e e ence genome sequence ound mo e han 120 ma ches be ween 73% and 80% iden i y, con aining mul iple s e ches o ~10 n pe ec iden i y, dispe sed o e all maize ch omosomes (Table1). Gi en hei numbe s, i is highly likely ha hese BLAST ma ches co espond o membe s o he Copia supe amily in he maize genome, which comp ises ~425–485 Mb o he maize genome17, he uni e sal p esence o in eg ase in in ac , au onomous LTR e o ansposons. Fu he esea ch is equi ed whe he o eign DNA may induce ecombina ion h ough sequence simila i y, especially when he o eign DNA exis in high concen a ion o high copy numbe s. Discussion Ba ley is classi ied as wo- owed o six- owed based on la e al spikele size and e ili y. Two- owed ba ley has a cen al e ile spikele and wo in e ile la e al spikele s, and six- owed ba ley has h ee ully-de eloped e ile spikele s. V s4 is an o holog o he maize (Zea mays. L.) in lo escence a chi ec u e gene RAMOSA2 (RA2), which encodes a LOB-domain-con aining ansc ip ional egula o 18–20. V s4 con ols ow- ype a ia ion and modi ies in lo escence de elopmen in ba ley (Ho deum ulga e. L)6. Exp ession analyses o mRNAs by in si u hyb idi- za ion and mic oa ay analysis e ealed ha V s4 is exp essed e y ea ly du ing in lo escence de elopmen and con ols he ow- ype pa hway in ba ley h ough V s1, a nega i e egula o o la e al spikele e ili y. The p bs mu an was ob ained om Pudamai-2, which has he no mal V s4 gene and a wo- owed pheno ype, by soaking he ba ley in lo escence in maize genomic DNA solu ion. In he mu an , V s4 is dele ed h ough a ecombina ion be ween wo BARE e o ansposons on ei he side o he V s4 gene. LTR e o ansposons a e known o ecombine; he ecombina ion be ween wo LTRs o a single elemen , which esul s in dele ion o he in e nal domain o he e o ansposon and gene a es a solo LTR, has been s udied16. The p ocess o e - o ansposon eplica ion gene a es LTRs ha a e iden ical a he momen o in eg a ion21; he accumula ion o mu a ions in he LTRs a he neu al a e a e ha allows o he es ima ion o he age o he in eg a ed ele- men 22. Genome-wide analyses show ha he a e age hal -li e o a e o ansposon in he Copia supe amily, which includes BARE, is 859,000 yea s, o a a e o 1.16 × 10−6 e en s pe elemen pe gene a ion, in he g ass B achypodium dis achyon, which loses e o ansposons h ough ecombina ion ela i ely apidly23. In ba ley and o he plan s, he a e o solo LTR o ma ion a ies conside ably be ween e o ansposon amilies and also be ween ch omosomes and egions20,23,24. Analyses o he equency o ecombina ion e en s be ween in e nal e o ansposon domains, such as he gene a ed p bs epo ed he e, ha e no been made and a e di icul o iden- i y in he absence o no el pheno ypes. Recombina ion be ween he LTRs o wo di e en elemen s can gene a e a conca ena ed s uc u e comp ising wo in e nal domains lanking a single, ecombinan LTR, which esul s in he loss o he in e ening genomic sequence, including any gene ha happens o be he e. A quan i a i e PCR su ey o he ba ley genome o such s uc u es wi h h ee LTRs and wo in e nal domains showed ha hei p esence in abou 4.3 × 103 copies pe haploid genome23. While his indica es he po en ial o gene loss h ough ecombina ion o e o ansposons lanking a gene, especially gi en ha he gene islands25 a e lanked by e o ansposon “seas” which inc eases he Mo ex-s a Mo ex-end Leng h Highes pe cen iden i y Highes e alue Hi numbe De ail alignmen esul s 115345 16193 848 73.01 1.484e-75 926 TableS2, Fig.S3 43155 43887 732 216423 18000 1577 72.61 2.035e-121 1019 TableS3, Fig.S4 44227 45804 1577 318046 18087 41 97.62 3.38E-12 879 TableS4, Fig.S5 45850 45891 41 4 27390 27543 153 81.1 5.543e-169 35 TableS5, Fig.S6 5 25868 26648 780 88.2 3.106e-45 1 TableS6, Fig.S7 Table 1. Sequence alignmen o he Mo ex dele ion wi h he maize genome sequence. www.na u e.com/scien i ic epo s/ 6 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 in e ening dis ance be ween ecombining e o ansposon sequences and appea s o be co ela ed wi h dec eas- ing ecombina ion equency20. The ques ion a ises as o whe he he maize DNA soaking p ocedu e is connec ed o he ecombina ion ha gene a ed he p bs mu an . Ou p ocedu e and he o he me hods in oduce o eign DNA in o he megagame o- phy e be o e e iliza ion26,27. Whe he o no any o eign DNA is in eg a ed, he p esence o ex a ch omosomal o cy oplasmic DNA igge s a ange o de ense esponses in animals28,29, media ed by DNA ecogni ion by p o- eins including STING (also called MITA, MPYS, TMEM173, o ERIS)30, speci ic oll-like ecep o s (TLRs)31, Z-DNA binding p o eins (ZBP-1, DLM-1, o DAI)32, and M e11 (meio ic ecombina ion 11)33. M e11 is pa icu- la ly in iguing because, oge he wi h RAD50 and NBS1, is a pa o he MRN complex and has been shown o play a i al ole in double-s and b eak (DSB) epai 34 in plan s, which is an in e media e s ep in ecombina ion. The maize genome con ains 404,000 Copia supe amily e o ansposons35,36; he in eg ase domains o hese a e e y simila o he in eg ase co e domain o BARE ha unde wen ecombina ion in p bs. We specula e whe he he homologous maize and ba ley in eg ase sequences may ha e in e ac ed wi h each o he , media - ing he ecombina ion. Exp ession analyses, in si u hyb idiza ion and mic oa ays e ealed ha V s4 is ac i ely exp essed du ing in lo escence de elopmen 6, co esponding o he s age a which he ba ley in lo escence was soaked in maize genomic DNA o gene a e he p bs mu an . Due o i s ansc ip ional ac i i y, his egion is likely o ha e an open ch oma in con o ma ion, which could p o ide an oppo uni y o maize DNA o in e ac a he BARE in eg ase domains and p omo e he ecombina ion. The high concen a ion o conse ed e o ansposon sequences would make binding and ecombina ion in a e o ansposon sequence mo e likely han elsewhe e. While ecombina ions be ween endogenous e o i uses (ERVs)—which a e s uc u ally iden ical o LTR e - o ansposons—ha e caused genic dele ions h ough ecombina ion37, o ou knowledge, he e has no been an ea lie demons a ion o his in plan s. Ho izon al gene ans e (HGT) is well documen ed in p oka yo ic genome e olu ion. I is ela i ely clea ha he e a e se e al HGT pa hways, including ans o ma ion, conjuga ion, and ansduc ion. In euka yo es, di ec DNA exchanges may occu du ing g a ing38, symbiosis39,40, pa asi ism41, pa hogenesis42, and epiphy e o en ophy e43. Some ec o s, such as pollen43, ungi44, bac e ia45, i uses46, plasmids38, insec s47 and anspos- ons43, may also be in ol ed in HGT. T ansposable elemen s (TEs) ha e been ecognized as impo an ec o s o he ho izon al mo emen o genes be ween euka yo ic genomes48–51. T ansposons, wi h hei inhe en abil- i y o mobilize, can p oli e a e and in eg a e in o genomic DNA and gene a e HGTs wi h ease52. T ansposons ha e also cap u ed and ansduced genomic DNA sequences in bo h Daphnia pulex53 and D osophila species52. The ans e o Mu-like ansposons be ween Se a ia and ice has been documen ed48,54. LTR e o ansposons can p oduce i us-like pa icles, which may wo k as mo e equen ec o s o HGT52,55. Such cases ha e been demons a ed in LTR- e o ansposon RIRE1 wi hin he genus O yza56 and he LTR- e o ansposon Rou e66 in Poaceae50. Wi h he inc easing a ailabili y o euka yo ic genome sequences, mo e e idence will be a ailable ha plan s a e also likely o unde go HGT. Howe e , he esul s ha e been based on incong uences in molecula phylogene ic ees. On he o he hand, he e a e nume ous epo s in he li e a u e ha ha e di ec ly in oduced o eign DNA by injec ing exogenous DNA o di ec ly DNA soaking o pollen ube pa hway in o ice, ba ley, whea , so ghum, maize, co on, oa s, ye, cucumbe , pumpkin, kidney bean and soybean o c ea e new gene ic a ia ions (Supplemen a yTable7). RAPD, AFLP and SSR molecula ma ke s ha e been used o es DNA ans- e s be ween species in se e al s udies57–59. Howe e , no s udy has demons a ed how he exogenous DNA causes gene ic a ia ion in o he species. Ou s udy p o ides p elimina y e idence ha LTR- e o ansposon-media ed gene dele ion/inse ion may play a ole in di ec gene ans e be ween di e en species. In addi ion o ac as po en ial ec o s o ho izon al gene ans e , ansposable elemen s a e also esponsi e and suscep ible o en i onmen al changes. I is well documen ed ha s esses could ac i a e TE o gene a e new gene ic di e si y60. This is especially ue o LTR- e o ansposons as he LTR is su icien in i sel o ac i a e TE ansc ip ion in esponse o s ess. I is possible ha soaking he ba ley spike in he maize DNA solu ion c ea ed s ess condi ions o he de eloping spikele s, which ac i a ed LTR-TE media ed ecombina ion. In his scena io, he maize genomic DNA may be no essen ial o he mu a ion. Fu he esea ch is equi ed o es his assump- ion by soaking he de eloping ba ley spikes in wa e o sal solu ion o p o ide simila s esses o iden i ica ion o new mu an s. Ma e ials and Me hods Plan ma e ials. A poly- ow b anched spike (p bs) ba ley mu an was ob ained by soaking a wo- owed ba - ley in lo escence (c . Pudamai-2) in maize genomic DNA solu ion61. The me hod ollowed ha desc ibed ea lie o whea 62. Flowe ing ba ley spikes we e soaked in o al maize DNA a 1.6 ug/ul in 0.1 × SSC o 24 hou s. A e soaking, he head was mo ed om he solu ion and ai -d ied unde ambien condi ions. Plan s we e sel -polli- na ed and seeds ha es ed. The mu an was iden i ied a lowe ing o he nex gene a ion plan s. Gene ic mapping was conduc ed in wo popula ions: one ecombinan inb ed line (RIL, F2:6) popula ion con- sis ing o 207 plan s de i ed om a c oss be ween he p bs mu an and a six- owed ba ley cul i a Kunlun 12, and an F2 popula ion consis ing o 285 spike mu an plan s de i ed om a c oss be ween he p bs mu an and a six- owed ba ley cul i a Zangqing 320. The p bs mu an , RIL 11R258-95 wi h a b anched spike pheno ype, Pudamai-2, and a . Mo ex we e used o DNA sequence analysis. Genomic DNA ex ac ion and geno ype analysis. Genomic DNA was ex ac ed om lea es o indi- idual plan s and hei pa en s using a modi ied CTAB me hod63. DNA samples we e quan i ied using a Unican UV300 UV/Vis spec ome e (The mo Elec on Co po a ion, Camb idge, UK), and hen adjus ed o 25 ng/μ l. Because a DQ327702 ma ke associa ed wi h he mu an is closely linked wi h he V s4 gene6, new molecula ma ke s Cbic43 and Cbic44 we e designed a ound he V s4 gene using he ba ley genome sequence14 om he IPK Ba ley BLAST se e (h p://webblas .ipk-ga e sleben.de/ba ley/ i oblas .php). P ime pai s speci ic o he www.na u e.com/scien i ic epo s/ 7 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 V s4 gene (AS12, AS34, and AS56)6 we e designed o V s4 haplo ype analysis. P ime s we e syn hesized by Shanghai Sunny Bio echnology (Shanghai, China). PCR eac ions we e pe o med in 10 μ L olumes con ain- ing app oxima ely 25 ng genomic DNA, 0.2 μ M o each p ime , and 5 μ L 2 × Taq Mas e Mix (Gene Solu ion, Shanghai, China) using he ollowing p og am: 94 °C o 3 min, 32 cycles o 94 °C o 30 sec, 55 °C o 45 sec, 72 °C o 1 min, and 72 °C o 5 min. PCR p oduc s we e sepa a ed on 8% polyac ylamide gels. Cloning o he dele ion in mu an p bs. BAC sequences we e iden i ied by blas ing he V s4 agains he In e na ional Ba ley Genome Sequencing Conso ium da abase (unpublished da a). PCR p ime pai s we e designed a 2 kb in e als in he egion nea he V s4 locus using he P ime -Blas ool (h p://www.ncbi.nlm.nih. go / ools/p ime -blas /). PCR eac ions we e desc ibed as abo e. Annealing empe a u es we e op imized o each p ime pai (Supplemen a yTable1). PCR p oduc s we e examined by elec opho esis on 1% aga ose gels. Long- ange PCRs we e pe o med in 50 μ L eac ions con aining 1 × bu e , 5 μ L empla e DNA, 0.4 μ M o each p ime , 400 μ M each deoxy ibonucleo ide, and 2.5 U LA Taq DNA polyme ase (Taka a, Dalian, China) using he ollowing p og am: 94 °C o 3 min, 32 cycles o 98 °C o 10 sec, 68 °C o 15 min, and 68 °C o 20 min. PCR p od- uc s we e examined by elec opho esis using a 0.8% aga ose gel, analyzed by Bio-Red Quan i y One gel image analysis sys em and sequenced by Shanghai RuiDi Biological Technology (Shanghai, China). Quan i a i e RT-PCR. RNA was ex ac ed om imma u e spikes a lemma p imo dium s age o he p bs mu an and wild pa en Pudamai-2 using Spin Column Plan o al RNA Pu i ica ion Ki (Sanggon Bio ech (Shanghai) Co.,L d). cDNA was p epa ed om 1 ug RNA using AMV Fi s S and cDNA Syn hesis Ki (Sanggon Bio ech (Shanghai) Co., L d). qPCR eac ions we e pe o med using SYBR G een (SG Fas qPCR Mas e Mix (HighRox), BBI ) and he Applied Biosys ems S epone plus Real- ime PCR Sys em. The Real- ime PCR assays we e pe o med in iplica e o each cDNA sample. V s46, V s14 and H Ac in6 p ime sequences used o quan i- a i e RT-PCR. The H Ac in gene was used as e e ence gene o no maliza ion. Sequence analysis. Alignmen s o mu an p bs and ba ley genomic sequences we e cons uc ed using MEGA 6.014 and BLASTN 2.3.0+ 64. The p edic ion o ansposable elemen s was iden i ied h ough LTR Finde 1.05 (h p:// li e. udan.edu.cn/l _ inde /)65 and BLAST66 agains he T i iceae Repe i i e Elemen (TREP) da a- base (h p://whea .pw.usda.go /GG2/blas .sh ml). Sea ches o sequence homology o maize was conduc ed wi h MaizeGDB agains he sequence da abase B73 Re Gen_ 3 (MGSC), using he BLAST p og am BLASTN63 wi h an E- alue cu o < 1e-50. Re e ences 1. Dous , A. N. & Kellogg, E. A. In lo escence di e si ica ion in he panicoid “b is le g ass” clade (Paniceae, Poaceae): e idence om molecula phylogenies and de elopmen al mo phology. Am. J. Bo . 89, 1203–1222 (2002). 2. S eeni asulu, N. & Schnu busch, T. A gene ic playg ound o enhancing g ain numbe in ce eals. T ends Plan Sci. 17, 91–101 (2012). 3. Fo s e , B. P. e al. The ba ley phy ome . Ann. Bo . 100, 725–733 (2007). 4. Koma suda, T. e al. Six- owed ba ley o igina ed om a mu a ion in a homeodomain-leucine zippe I-class homeobox gene. P oc. Na l. Acad. Sci. USA 104, 1424–1429 (2007). 5. Ramsay, L. e al. INTERMEDIUM-C, a modi ie o la e al spikele e ili y in ba ley, is an o holog o he maize domes ica ion gene TEOSINTE BRANCHED 1. Na . Gene . 43, 169–172 (2011). 6. Koppolu, R. e al. Six- owed spike4 (V s4) con ols spikele de e minacy and ow- ype in ba ley. P oc. Na l. Acad. Sci. USA 110, 13198–13203 (2013). 7. Lundq is , U. & Lundq is , A. Induced in e medium mu an s in ba ley: o igin, mo phology and inhe i ance. He edi as. 108, 13–26 (1988). 8. Lundq is , U., F anckowiak, J. D. & Konishi, T. New and e ised desc ip ions o ba ley genes. Ba ley Gene . Newsl. 26, 22–516 (1997). 9. Liu, S. e al. S udies on inhe i ance and spike cha ac e s o poly- ow-and-b anched spike mu an in ba ley. Fujian Science and Technology o Rice and Whea . 18, 37–39 (2000). 10. Ji, H., Chen, Q. & Lin, X. S udy on poly- ow ba ley wi h mul i-b anches o spikele s eme ged by di ec ly o DNAs. Jou nal o Fujian Ag icul u e Uni e si y 24, 9–13 (1995). 11. Huang, B., Wu, W., Liu, S. & Huang, Z. Gene ic Analysis on Poly- ow-and-b anched Spike Mu an in Ba ley. He edi as (Beijing) 26, 903–906 (2004). 12. Shang, Y. e al. Cha ac e iza ion and mapping o a P bs gene con olling spike de elopmen in Ho deum ulga e L. Genes Genomics 36, 275–282 (2014). 13. Suoniemi, A., Tanskanen, J., Pen ikainen, O., Johnson, M. S. & Schulman, A. H. The co e domain o e o ansposon in eg ase in Ho deum: p edic ed s uc u e and e olu ion. Mol. Biol. E ol. 15, 1135–1144 (1998). 14. Tamu a, K., S eche , G., Pe e son, D., Filipski, A. & Kuma , S. MEGA6: Molecula E olu iona y Gene ics Analysis e sion 6.0. Mol. Biol. E ol. 30, 2725–2729 (2013). 15. K ishnan, L. & Engelman, A. Re o i al in eg ase p o eins and HIV-1 DNA in eg a ion. J. Biol. Chem. 287, 40858–40866 (2012). 16. Vi e, C. & Panaud, O. Fo ma ion o solo-LTRs h ough unequal homologous ecombina ion coun e balances ampli ica ions o LTR e o ansposons in ice O yza sa i a L. Mol. Biol. E ol. 20, 528–540 (2003). 17. Es ep, M. C., DeBa y, J. D. & Benne zen, J. L. The dynamics o LTR e o ansposon accumula ion ac oss 25 million yea s o panicoid g ass e olu ion. He edi y (Edinb). 110, 194–204 (2013). 18. Bo i i, E. e al. Ramosa2 Encodes a LATERAL ORGAN BOUNDARY Domain P o ein Tha De e mines he Fa e o S em Cells in B anch Me is ems o Maize. Plan Cell 18, 574–585 (2006). 19. Shi asu, K., Schulman, A. H., Lahaye, T. & Schulze-Le e , P. A con iguous 66-kb ba ley DNA sequence p o ides e idence o e e sible genome expansion. Genome Res. 10, 908–915 (2000). 20. Baidou i, M. E. & Panaud, O. Compa a i e Genomic Paleon ology Ac oss Plan Kingdom Re eals The Dynamics O TE-d i en Genome E olu ion. Genome Biol E ol. 5, 954–965 (2013). 21. Schulman, A. H. Re o ansposon eplica ion in plan s. Cu Opin Vi ol. 3, 604–614 (2013). 22. SanMiguel, P., Gau , B. S., Tikhono , A., Nakajima, Y. & Benne zen, J. L. The paleon ology o in e gene e o ansposons o maize. Na . Gene . 20, 43–45 (1998). 23. Vogel, J. P. e al. Genome sequencing and analysis o he model g ass B achypodium dis achyon. Na u e. 463, 763–768 (2010). 24. Maye , K. F. e al. A physical, gene ic and unc ional sequence assembly o he ba ley genome. Na u e 491, 711–716 (2012). www.na u e.com/scien i ic epo s/ 8 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 25. Liu, R. e al. A GeneT ek analysis o he maize genome. P oc. Na l. Acad. Sci. USA 104, 11844–11849 (2007). 26. Zhou, G. Y. e al. In oduc ion o exogenous DNA in o co on emb yos. Me h. Enzymol. 101, 433–481 (1983). 27. Peña, A. D. L., Lö z, H. & Schell, J. T ansgenic ye plan s ob ained by injec ing DNA in o young lo al ille s. Na u e. 325, 274–276 (1987). 28. Ba be , G. N. STING-dependen cy osolic DNA sensing pa hways. T ends Immunol. 35, 88–93 (2014). 29. Ma, Z. e al. Modula ion o he cGAS-STING DNA sensing pa hway by gammahe pes i uses. P oc. Na l. Acad. Sci. USA 112, 4306–4315 (2015). 30. Abe, T. e al. STING Recogni ion o Cy oplasmic DNA Ins iga es Cellula De ense. Mol. Cell. 50, 5–15 (2013). 31. Kawai, T. & Aki a, S. Toll-like Recep o s and Thei C oss alk wi h O he Inna e Recep o s in In ec ion and Immuni y. Immuni y 34, 637–650 (2011). 32. Takaoka, A. e al. DAI (DLM-1/ZBP1) is a cy osolic DNA senso and an ac i a o o inna e immune esponse. Na u e 448, 501–505 (2007). 33. Kondo, T. e al. DNA damage senso MRE11 ecognizes cy osolic double-s anded DNA and induces ype I in e e on by egula ing STING a icking. P oc. Na l. Acad. Sci. USA 110, 2969–2974 (2013). 34. Samanic, I., C i anic, R., Simunic, J. & Puizina, J. A abidopsis haliana MRE11 is essen ial o ac i a ion o he cell cycle a es , ansc ip ional egula ion and he DNA epai upon he induc ion o double- s anded DNA b eaks. Plan Biol (S u g). 18, 681–694 (2016). 35. Tenaillon, M. I., Hu o d, M. B., Gau , B. S. & Rossiba a, J. Genome size and ansposable elemen con en as de e mined by high- h oughpu sequencing in maize and Zea luxu ians. Genome Biol E ol. 3, 219–229 (2011). 36. Schnable, P. S. e al. The B73 maize genome: complexi y, di e si y, and dynamics. Science. 326, 1112–1115 (2009). 37. Shu a iko , A. e al. Recu en HERV-H-media ed 3q13.2-q13.31 dele ions cause a synd ome o hypo onia and mo o , language, and cogni i e delays. Hum. Mu a . 34, 1415–1423 (2013). 38. S egemann, S. & Bock, R. Exchange o gene ic ma e ial be ween cells in plan issue g a s. Science 324, 649–651 (2009). 39. Finan, T. M. E ol ing insigh s: symbiosis islands and ho izon al gene ans e . J. Bac e iol. 184, 2855–2856 (2002). 40. Bu ge , G. & Lang, B. F. Pa allels in genome e olu ion in mi ochond ia and bac e ial symbion s. IUBMB Li e 55, 205–212 (2003). 41. Yoshida, S., Ma uyama, S., Nozaki, H. & Shi asu, K. Ho izon al gene ans e by he pa asi ic plan S iga he mon hica. Science 328, 1128–1128 (2010). 42. Sha p, A. J. e al. Disco e y o p e iously uniden i ied genomic diso de s om he duplica ion a chi ec u e o he human genome. Na . Gene . 38, 1038–1042 (2006). 43. Bock, R. The gi e-and- ake o DNA: ho izon al gene ans e in plan s. T ends Plan Sci. 15, 11–22 (2010). 44. Richa ds, T. A. e al. Phylogenomic analysis demons a es a pa e n o a e and ancien ho izon al gene ans e be ween plan s and ungi. Plan Cell. 21, 1897–1911 (2009). 45. B oo hae s, W. e al. Gene ans e o plan s by di e se species o bac e ia. Na u e 433, 629–633 (2005). 46. Hull, R., Ha pe , G. & Lockha , B. Vi al sequences in eg a ed in o plan genomes. Annu Re Phy opa hol. 5, 362–365 (2000). 47. Die e ich, C. e al. The P is ionchus paci icus genome p o ides a unique pe spec i e on nema ode li es yle and pa asi ism. Na . Gene . 40, 1193–1198 (2008). 48. Diao, Y. e al. Nex -gene a ion sequencing e eals ecen ho izon al ans e o a DNA ansposon be ween di e gen mosqui oes. PloS One 6, e16743 (2011). 49. No ick, P., Smi h, J., Ray, D. & Boissino , S. Independen and pa allel la e al ans e o DNA ansposons in e apod genomes. Gene. 449, 85–94 (2010). 50. Roulin, A. e al. Whole genome su eys o ice, maize and so ghum e eal mul iple ho izon al ans e s o he LTR- e o ansposon Rou e66 in Poaceae. BMC E ol. Biol. 9, 58 (2009). 51. So mache a, I. e al. Ve ical e olu ion and ho izon al ans e o CR1 non-LTR e o ansposons and Tc1/ma ine DNA ansposons in Lepidop e a species. Mol. Biol. E ol. 9, 3685–3702 (2012). 52. Lo e o, E. L. S., Ca a e o, C. M. A. & Capy, P. Re isi ing ho izon al ans e o ansposable elemen s in D osophila. He edi y. 100, 545–554 (2008). 53. Schaack, S., Choi, E., Lynch, M. & P i ham, E. J. DNA ansposons and he ole o ecombina ion in mu a ion accumula ion in Daphnia pulex. Genome Biol. 11, R46 (2010). 54. Diao, X., F eeling, M. & Lisch, D. Ho izon al ans e o a plan ansposon. PLoS Biol. 4, 119 (2006). 55. Sil a, J. C., Lo e o, E. L. & Cla k, J. B. Fac o s ha a ec he ho izon al ans e o ansposable elemen s. Cu Issues Mol Biol. 6, 57–71 (2004). 56. Roulin, A., Piegu, B., Wing, R. A. & Panaud, O. E idence o mul iple ho izon al ans e s o he long e minal epea e o ansposon RIRE1 wi hin he genus O yza. Plan J. 53, 950–959 (2008). 57. Luo, H., Zhong, B., Yang, Z., Li, Y. & He, G. The SSR molecula e idence o ice ans o ma ion ia pollen ube pa hway. Mol. Plan B eed. 2, 501–505 (2004). 58. Liu, P. & Kang Z. Oa (A ena sa i a L) exogenous DNA in oduc ion in o common whea and PAPD analysis. Ag icul u al Resea ch in he A id A eas. 24, 100–103 (2006). 59. Wang, S. e al. Molecula e i ica ion o DNA low om wild ice (O. minu a) o cul i a ed ice. Scien ia Ag icul u a Sinica. 39, 2170–2177 (2006). 60. Casacube a, E. & Gonzalez, J. The impac o ansposable elemen s in en i onmen al adap a ion. Molecula Ecology 22, 1503–1517 (2013). 61. Huang, B. & Wu, W. Mapping o Mu an Gene p bs Con olling Poly-Row-and-B anched Spike in Ba ley (Ho deum ulga e L.). Ag ic. Sci. China 10, 1501–1505 (2011). 62. Bai, F. e al. C ea ing a New Whea S ain o Ea ly Ma u i y and Ano he One o Good Dwa Quali y by In oducing Exogeous Maize Nuclea DNA. Ac a Ag onomica Sinica. 25, 260–264 (1999). 63. Lu, Y. & Zheng, K. A simple me hod o isola ion o ice DNA. Chinese Jou nal o Rice Science 6, 47–48 (1992). 64. Zhang, Z., Schwa z, S., Wagne , L. & Mille , W. A g eedy algo i hm o aligning DNA sequences. J. Compu . Biol. 7, 203–214 (2000). 65. Xu, Z. & Wang, H. LTR_FINDER: an e icien ool o he p edic ion o ull-leng h LTR e o ansposons. Nucleic Acids Res. 35, 265–268 (2007). 66. Al schul, S. F. e al. Gapped BLAST and PSI-BLAST: a new gene a ion o p o ein da abase sea ch p og ams. Nucleic Acids Res. 25, 3389–3402 (1997). Acknowledgemen s We a e g a e ul o P o esso Jing Zhang, Ins i u e o C op Ge mplasm Resou ces, Chinese Academy o Ag icul u al Science, o p o iding he mu an p bs, and o D Nils S ein Leibniz Ins i u e o Plan Gene ics and C op Plan Resea ch (IPK) o accessing he un-published ba ley BAC sequence. This wo k was pa ially suppo ed by he Na u al Science Founda ion o Zhejiang P o ince (LY15C130004), China Ag icul u e Resea ch Sys em (CARS-05), he Na ional Na u e Science Founda ion o China (31501309) and he Academy o Finland, P ojec 266430. www.na u e.com/scien i ic epo s/ 9 Scien i ic RepoR s | 7:43766 | DOI: 10.1038/s ep43766 Au ho Con ibu ions C.L. and J.Y. concei ed, designed and supe ised he s udy; Y.S. and F.Y. a e p incipal in es iga o s; A.H.S., F.Y.W.Z. and Y.J. da a analysis; J.Z., J.W., W.H. pheno yping; N.S., X.Z. and Q.J. DNA sequencing; Y.S., F.Y. A.S. and C.L. w o e he pape . Addi ional In o ma ion Supplemen a y in o ma ion accompanies his pape a h p://www.na u e.com/s ep Compe ing In e es s: The au ho s decla e no compe ing inancial in e es s. How o ci e his a icle: Shang, Y. e al. Gene Dele ion in Ba ley Media ed by LTR- e o ansposon BARE. Sci. Rep. 7, 43766; doi: 10.1038/s ep43766 (2017). Publishe 's no e: Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. This wo k is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s C ea i e Commons license, unless indica ed o he wise in he c edi line; i he ma e ial is no included unde he C ea i e Commons license, use s will need o ob ain pe mission om he license holde o ep oduce he ma e ial. To iew a copy o his license, isi h p://c ea i ecommons.o g/licenses/by/4.0/ © The Au ho (s) 2017