Transposable elements in a marginal plant population: temporal fluctuations provide new insights into genome evolution of wild diploid wheat

RESEARCH Open Access T ansposable elemen s in a ma ginal plan popula ion: empo al luc ua ions p o ide new insigh s in o genome e olu ion o wild diploid whea Alexande Belyaye 1* , Ruslan Kalenda 2 , Leonid B odsky 1 , E ia a Ne o 1 , Alan H Schulman 2,3* , Olga Raskina 1 Abs ac Backg ound: How new o ms a ise in na u e has engaged e olu iona y biologis s since Da win’s seminal ea ise on he o igin o species. T ansposable elemen s (TEs) may be among he mos impo an in e nal sou ces o in aspeci ic a iabili y. Thus, we aimed o explo e he empo al dynamics o se e al TEs in indi idual geno ypes om a small, ma ginal popula ion o Aegilops spel oides. A diploid c oss-pollina ed g ass species, i is a wild ela i e o he a ious whea species known o hei la ge genome sizes con ibu ed by an ex ao dina y numbe o TEs, pa icula ly long e minal epea (LTR) e o ansposons. The popula ion is cha ac e ized by high he e omo phy and possesses a wide spec um o ch omosomal abno mali ies including supe nume a y ch omosomes, he e ozygosi y o ansloca ions, and a iabili y in he ch omosomal posi ion o numbe o 45S and 5S ibosomal DNA ( DNA) si es. We p opose ha a iabili y on he mo phological and ch omosomal le els may be linked o a iabili y a he molecula le el and pa icula ly in TE p oli e a ion. Resul s: Signi ican empo al luc ua ion in he copy numbe o TEs was de ec ed when p ocesses ha ake place in small, ma ginal popula ions we e simula ed. I is known ha unde c i ical ex e nal condi ions, ou c ossing plan s e y o en ansi o sel -pollina ion. Thus, h ee mo phologically di e en geno ypes wi h ch omosomal abe a ions we e aken om a wild popula ion o Ae. spel oides, and he dynamics o he TE complex aced h ough h ee ounds o sel ing. I was disco e ed ha : (i) a ious amilies o TEs a y emendously in copy numbe be ween indi iduals om he same popula ion and he sel ed p ogenies; (ii) he luc ua ions in copy numbe a e TE- amily speci ic; (iii) he e is a g ea di e ence in TE copy numbe expansion o con ac ion be ween game ophy es and spo ophy es; and (i ) a small pe cen age o TEs ha inc ease in copy numbe can ac ually inse a no el loca ions and could se e as a bona ide mu agen. Conclusions: We hypo hesize ha TE dynamics could p omo e o in ensi y mo phological and ka yo ypical changes, some o which may be po en ially impo an o he p ocess o mic oe olu ion, and allow species wi h plas ic genomes o su i e as new o ms o e en species in imes o apid clima ic change. Backg ound Popula ions a e gene ally iewed as he elemen a y e o- lu iona y uni [1,2]. A popula ion exis s as an in eg a ion o indi iduals in ime and space ha can a y o e a se o successi e gene a ions. Unde he in luence o spon- aneous mu a ions, each popula ion becomes he e ogeneous in i s gene ic s uc u e o e ime. Thus, a popula ion comp ises a mix o di e en geno ypes e en i i s indi iduals a e mo e o less pheno ypically simila . Unde in ensi e p essu e om a pa icula (usually en i onmen al) ac o , a common phenomenon in ma - ginal popula ions, a shi in he geno ypic s uc u e o he popula ion can occu . Such a shi may be ega ded as he elemen a y e en o mic oe olu ion. T ansposable elemen s (TEs) may be among he mos impo an in e nal sou ces o geno ypic popula ion * Co espondence: [email p o ec ed]ai a.ac.il; [email p o ec ed] 1 Ins i u e o E olu ion, Uni e si y o Hai a, Moun Ca mel, Hai a, Is ael 2 MTT/BI Plan Genomics Labo a o y, Ins i u e o Bio echnology, Viikki Biocen e , Uni e si y o Helsinki, Helsinki, Finland Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 © 2010 Belyaye e al; licensee BioMed Cen al L d. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/2.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. change as a esul o hei abili y o c ea e mu a ions, al e gene exp ession, and p omo e ch omosomal abe - a ions [3-5]. These ancien , ubiqui ous, and dynamic componen s o euka yo ic genomes comp ise up o 80% o he la ge genomes o ce eals [6,7]. Da a accumula ed o e he las 30 yea s sugges ha TEs can ha e majo e ec s on genome o ganiza ion and unc ion [4,6,8,9]. Ne e heless, i is s ill unclea whe he and o wha deg ee he TEs con ibu e o e olu iona ily signi ican shi s in he geno ypic s uc u e o popula ions. Popula- ion gene ic heo y assumes ha he dynamics o ans- posable elemen s in na u al popula ions e lec a balance be ween he endency o hese elemen s o inc ease h ough ansposi ion and hei emo al h ough na u al selec ion ac ing agains indi iduals wi h a high elemen copy numbe [10,11]. Se e al s udies om di e en bio- logical sys ems indica e ha he pa e ns o a single TE amily can a y in aspeci ically [12-15] and empo ally [16]. Examina ion o he in aspeci ic dynamics o TEs, especially in ma ginal popula ions whe e mic oe olu- iona y p ocesses a e in ensi ied [2], could conside ably con ibu e o he unde s anding o key biological e en s such as specia ion. One o he mos e ec i e ways o unde s and he dynamics o he TEs o e ime and space may be an ecological app oach, whe eby se e al gene a ions o plan s om na u al popula ions a e in es iga ed. He e, we ha e chosen a small, ma ginal and h ea ened popu- la ion o Aegilops spel oides (2n = 2× = 14) as a model o explo a ion o he dynamics o se e al TE amilies belonging o Classes I and II ( espec i ely mo ing ia RNA and DNA in e media es). The species is a wild ela i e o he a ious whea species known o hei la ge genome sizes [17,18] con ibu ed o by an ex ao - dina y numbe o TEs, pa icula ly long e minal epea (LTR) e o ansposons. Ae. spel oides was p oposed o be he closes o he wild diploid p ogeni o o he G genomes and B genomes o polyploid whea [19-21]. The s udy popula ion is loca ed on he wes e n banks o he Kishon Ri e (Hai a Bay a ea, Is ael). The popula ion is cha ac e ized by high he e omo phy and possesses a wide spec um o ch omosomal abno mali ies, including supe nume a y ch omosomes, he e ozygosi y o ans- loca ions, and a iabili y in he ch omosomal posi ion o numbe o 45S and 5S ibosomal DNA ( DNA) si es [22,23]. We p opose ha a iabili y in mo phological and ch omosomal le els may poin o a iabili y a he molecula le el and pa icula ly in TE p oli e a ion. Thus we aimed o de ec and e alua e di e ences in TE copy numbe s be ween indi iduals om he Kishon popula ion and o ace inhe i ance o hese de ia ions in successi e gene a ions. The b oade goal o he s udy was o examine whe he TE dynamics could be asso- cia ed wi h mo phological o ka yo ypical changes, some o which a e po en ially impo an o he e olu iona y p ocess. Resul s Expe imen al design Ae. spel oides is p edominan ly a c oss-pollina ed bu sel -compa ible species [24]. Th ee o iginal plan s we e selec ed om a small (app oxima ely 100 m 2 ), ma ginal, deg ade popula ion o Ae. spel oides (see Me hods). Each selec ed o iginal geno ype ep esen s h ee g oups o p e iously in es iga ed plan s ( i e o se en indi idual o iginal spikes in each g oup), which ha e been clus- e ed due o hei mo phological and cy ogene ic simi- la i y, including spike mo phology, B ch omosome exis ence, appea ance o addi ional 5S DNA ch omoso- mal clus e s, and speci ic ch omosomal ea angemen s [22,23]. The p ogeny om each geno ype we e ob ained in h ee ounds o sel ing. We simula ed he si ua ion in na u e whe e, in ma ginal popula ions unde c i ical ex e nal condi ions, ou c ossing plan s and pa icula ly Ae. spel oides e y o en ansi o sel -pollina ion [24-26]. The copy numbe s o se e alTE amilieswe e de e mined by quan i a i e polyme ase chain eac ion (qPCR) o each o iginal geno ype and i s o sp ing. qPCR da a we e e i ied by do blo analysis. The ans- posi ional ac i i ies o TEs we e in e ed by in e - e o- ansposon ampli ied polymo phism (IRAP) e o ansposon display [27]. Gene a ional changes in he copy numbe o TE amilies we e compa ed wi h hose o he non-mobile, highly epe i i e andem epea Spel 52 and he 5S ibosomal RNA ( RNA) genes ( DNA). Dynamics o TE in successi e gene a ions The copy numbe s o TE amilies ha we in es iga ed a e p esen ed in Addi ional ile 1 and in g aphic o m in Figu e 1a. The changes in TE copy numbe in a se o successi e gene a ions was measu ed by wo di e en me hods: qPCR and do blo . These p oduced consis en esul s. The i s salien poin is ha gene a i e issues, in mos cases, possess highe numbe s o TEs han do ege a i e issues. This obse a ion is ue o mos TEs excep hose ela ed o he A hila amily (Sab ina and Wham). Fo example, he numbe o WIS2 e o ansposons in he spikes o G13 S1 exceeded ha in lea es by 59%, o app oxima ely 18,000 copies. The copy numbe s o bo h he ibe-speci ic Spel 52 andem epea s and he 5S RNA genes we e also gene ally highe in spikes han in lea es (Addi ional ile 1, Figu e 1a). The second ema kable ea u e is ha he copy num- be s o all in es iga ed TEs a ied signi ican ly o e h ee successi e gene a ions, and ha each geno ype possessed indi idual TE dynamics o e his pe iod. The Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 2 o 16 TE copy numbe s dec eased o inc eased signi ican ly, wi h a ise in one gene a ion ollowed by a d op in he nex and ice e sa in an oscilla o y ashion. In gene a- i e issues, hese oscilla ions displayed much highe ampli udes. Pa icula TEs dec eased up o 23% om hei ini ial quan i y, as in he case o Daniela e o an- sposons in G9 spikes whe e he copy numbe o his elemen d opped om app oxima ely 62,000 copies in S1 o app oxima ely 15,000 copies in S2 (Addi ional ile 1, Figu e 1a). We could no ace u he TE dynamics in his geno ype since he S2 gene a ion was s e ile. The la ges absolu e ise in TE numbe was documen ed also o Daniela in G13 spikes whe e he copy numbe ose om app oxima ely 18,000 copies in S2 o app oxi- ma ely 52,000 copies in S3. The la ges ela i e ise in TE abundance in one gene a ion was seen o he Sab- ina elemen , whe e in G14 i s numbe in spikes inc eased by 672% in S3 as compa ed o S2. A simila jump in his geno ype be ween S2 and S3 was obse ed also o WIS2,Wilma, and Wham. Geno ypes 9 and 13 we e dis inc by he ampli udes in hei numbe s o copies o pa icula TE amilies o e he gene a ions analyzed, bu he endencies we e he same. Common ends included a signi ican inc ease o he highly abundan e o ansposons, WIS2,Daniela, and Fa ima, in he gene a i e issues o he S1 Figu e 1 Copy numbe s and mo phology.(a) Dynamics o ansposable elemen (TE) copy numbe s in h ee sel -pollina ed gene a ions o h ee geno ypes om he Kishon popula ion o Aegilops spel oides (shown by lines). TE copy numbe s o he TS-84 popula ion we e used as con ols. TE copy numbe s o a ailable sibs in sel ing gene a ions a e shown by sepa a e do s. (b) Changes in spike mo phology in h ee sel - pollina ed gene a ions o h ee geno ypes om he Kishon popula ion o Ae. spel oides. Spike mo phology o plan s om he TS-84 popula ion was used as he con ol. Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 3 o 16 gene a ion, ollowed by a simila dec ease in hese ele- men s in he S2 gene a ion (Addi ional ile 1, Figu e 1a). En/Spm-like ansposons, Cassand a e o ansposons, and 5S RNA genes exhibi ed simila dynamics in hese geno ypes. Wilma e o ansposons did no change in he S1 gene a ion bu hen inc eased in numbe in S2. The copy numbe o Sukkula elemen s did no change in he S1 gene a ion, bu hen d opped in S2. Sab ina and Wham elemen s emained app oxima ely on he same le el as in he o iginal plan s. The TE dynamics in ege a i e issues we e app oxima ely he same as in gene a i e issues bu wi h less ampli ude. Fo echnical easons, we do no ha e measu emen s o he copy numbe o Spel 52 in he spikes o he o iginal plan s o G9, bu he es o he measu emen s showed a en- dency o his andem epea o dec ease in abundance in gene a i e and ege a i e issues o e he gene a ions examined. Geno ype 14 di e ed in TE copy numbe empo al dynamics om he wo p e iously desc ibed geno ypes. Cy ogene ic da a (see below) make i possible o p opose ha his geno ype was al eady sel -pollina ed o a leas one gene a ion. WIS2, Wilma, Sab ina, Wham, and Suk- kula elemen s demons a ed signi ican , successi e dec eases in copy numbe s in S1 and S2 ha we e hen ollowed by inc eases in S3. Daniela, En/Spm,andCas- sand a elemen s inc eased in copy numbe in S1 and successi ely dec eased in S2 and S3. Fa ima e o an- sposons exhibi ed copy numbe dynamics in G14 simila o hose seen in G9 and G13: inc eased abundance in S1, dec easing in S2, and again inc easing in S3. In G14, oo, he TE dynamics in ege a i e issues we e app oxi- ma ely he same as o gene a i e issues, bu wi h less ampli ude. The numbe o Spel 52 epea s, bo h in spikes and lea es, dec eased in S1, inc eased in S2, and dec eased in S3. In spi e o a signi ican inc ease o copy numbe in S2, he o e all endency was o educ ion. The 5S RNA genes successi ely inc eased in copy num- be in S1 and S2, and hen dec eased in S3. To explo e seg ega ion in he p ogenies o a single geno ype, TE copy numbe s in sibs om G9 and G13 geno ypes we e de e mined (Table S1 in Addi ional ile 2). Signi ican a ia ions o e h ee gene a ions o wo geno ypes we e obse ed. The ampli ude o TE abun- dance in sibs could no be explained by simple ch omo- some seg ega ion (see below) and, he e o e, poin s o mobile elemen ac i i y. A TE display me hod such as IRAP [27] would be expec ed o show polymo phisms consonan wi h la ge changes in TE copy numbe . He e, IRAP analyses we e conduc ed on DNA om spikes a he mic ospo ogen- esis s age, and showed a high le el o polymo phism om gene a ion o gene a ion (Figu e 2). G13 showed he mos unique bands, wi h e o ansposons WIS2 and Daniela p oducing he g ea es numbe . Va ious o ms o ecombina ion may, o cou se, b ing TEs su icien ly close ha a new IRAP band would appea in he absence o a new in eg a ion e en . Fo any gi en IRAP polymo phism, i s o igin as an in eg a- ion e en can be explici ly p o en only by iden i ying an emp y si e, om a plan line missing he IRAP p o- duc , o one o he wo TEs ha oge he se ed as he empla e. To ind e idence o indi idual ans o ma ion e en s, we cloned and sequenced 20 unique IRAP bands (Figu e 3). A leas hal o he new inse ion si es we e in epe i i e DNA, and he e o e unsui ed o iden i ying he o iginal emp y si es. O he 20, (Table S2 in Addi- ional ile 2), 4 o hem (a Daniela,aWis2,and2Suk- kula inse ions) we e in non- epe i i e si es ha could be iden i ied in a sequence da abase. The ad en o new, unique, ampli ied IRAP bands ha esul ed om new TE inse ions con i ms he exis ence o ansposi ionally ac i e TEs in he h ee geno ypes o Ae. spel oides. Mul i a ia e analyses The da a dis ibu ion in p inciple componen analysis (PCA) (Figu e 4a) demons a es he clea sepa a ion o lea es and spikes o each gene a ion, no malized pai - wise o each TE and non-TE epea in con ol plan s. The sepa a ion is along he i s p incipal componen (PC1), which co e s 62% o ini ial da a a iabili y. The PC1 (Figu e 4b) con as s a g oup o TEs (Sukkula, Fa ima, Cassand a, En/Spm, and Daniela) and wo non- TE sequences (Spel 52 and 5S DNA) e sus ano he g oup o TEs (Wilma, Sab ina, Wham,andWIS2). The e o e, he pa e ns o ela i e gene a ional copy numbe a ia ion o hese wo g oups o TEs ega ding hei numbe s in con ol plan s a e di e en in spikes and lea es. The i s g oup o TEs and bo h non-TE sequences show la ge luc ua ions in lea es o e suc- cessi e gene a ions, and he TEs o he second g oup show la ge changes in spikes. The second p incipal componen (PC2) ep esen s he a iabili y ha is independen o he lea -spike con as (Figu e 4b). When his componen is u he dissec ed by ype o epea (Figu e 4c), he issue-independen a iance shows opposi e ends o TE and non-TE epea s. This dicho omous end holds ac oss gene a- ions, geno ypes, and issues. We u he in es iga ed he na u e o his end o all nine TE amilies. Bo h aw copy numbe s and hose no malized by he con ol da a we e hus no malized by he numbe o one o he non-TE sequences (5S DNA) and by con ol plan s (TS-84) (Figu e 4d). Fo bo h ypes o da a no maliza- ion, he cen alized pa e ns o change in copy numbe o he TEs and o Spel 52 o e he h ee geno ypes and ac oss issues and gene a ions demons a e ha : (i) he copy numbe luc ua ion o non-TE sequence Spel Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 4 o 16 52 wi h espec o 5S DNA is insigni ican ac oss is- sues and gene a ions (b oken ed line) while luc ua ions o TE sequences a e high; (ii) he in e gene a ional pa - e n o copy numbe changes o all TEs is highly co e- la ed; (iii) he e is li le ela i e TE copy numbe a ia ion in lea es hough i is e y s ong in spikes. I shows a minimum in he S2 gene a ion, and a maxi- mum in he S3 gene a ion. In sum, he sepa a ion o TE and non-TE sequences in PC2 appea s ela ed o hei di e en pa e ns o copy numbe dynamics ac oss gen- e a ions in spikes. De ia ions o eal copy numbe s in lea es and spikes om he s ochas ic model The o iginal plan s o c oss-pollina ed Ae. spel oides a e he e ozygous o epe i i e DNA sequences, ch o- mosomal pa e ns, and ch omosomal ea angemen s (Figu es 5 and 6). In o de o sepa a e he ole o he p e-exis ing he e ozygosi y on he TE copy numbe in Figu e 2 In e - e o ansposon ampli ied polymo phism (IRAP) analyses o se e al ansposable elemen s (TEs) in he p ogeny o h ee geno ypes. Unique bands ha appea in S2 and a e inhe i ed in S3 a e shown by ed a ows. An example o he e ozygosi y displayed in he IRAP pa e n is shown in he blue squa e. An example o band loss in S2 and S3 is shown in he g een squa e. An example o band appea ance in S2 and S3 is shown in he yellow squa e. Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 5 o 16 he o iginal plan s om o he mechanisms ha can change TE amoun s in successi e gene a ions, such as ansposi ion, excision, o ecombina ional elimina ion, we c ea ed a ‘naï e’model in which he seg ega ion o pa en al ch omosomes in sel ed p ogenies was simu- la ed. The null hypo hesis implies ha : (i) c oss-polli- na ed pa en al plan s we e ini ially he e ozygous o he con en o TEs in he homologues; (ii) only he andom seg ega ion o homologues in meiosis du ing male and emale game ogenesis leads o a change in TE copy numbe in inb ed o sp ing. Fo modeling we ook only he G13 geno ype as he geno ype wi h a s able ch omosome numbe , and only he dynamics o Class I elemen s we e simula ed. We ca ied ou 100 ealiza ions o s ochas ic ime cou se simula ions o copy numbe s in spikes and lea es ac oss en gene a ions. Thus, he copy numbe s in lea es and spikes we e calcula ed in 100 simula ed siblings o each gene a ion o plan s, and hese 100 simula ed siblings om each o he i s h ee plan gene a ions o m he gene a ion-speci ic basic dis ibu ions o copy numbe s in lea es and spikes. The de ia ions o copy numbe s in eal siblings om he basic simula ed dis ibu ions, sepa a ely in lea es and spikes, es he null hypo hesis. The signi icance o de ia ions in each gene a ion was calcula ed by a - es and he non-pa ame ic Mann- Whi ney U es (Table S7 in Addi ional ile 2). The sig- ni icance esul s o wo es s a e in good mu ual conco - dance. Taking his in o conside a ion, he F- es s a is ics we e used o signi icance es ima ion o he cumula i e de ia ion in posi i e o nega i e di ec ions o eal da a om simula ed siblings ac oss h ee gene a- ions o each ype o TE sepa a ely. The F s a is ic is: FW Wmax min / whe e W max and W min a e ei he W p o W n depending on which one is la ge . The s a is ic W p is de ined as he sum o squa es ei he o eal - es alues, i - es is Figu e 3 Analysis o unique in e - e o ansposon ampli ied polymo phism (IRAP) bands.In hisexample,IRAPwasconduc edwi h p ime 2109, ma ching he Daniela e o ansposon long e minal epea (LTR) (Table S2 in Addi ional ile 2). The band was cu and sequenced. Two new p ime s we e designed o ma ch he sequence and he uniqueness o he band was checked on he se o DNA samples. This new inse ion is in epe i i e DNA (Table S2 in Addi ional ile 2). Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 6 o 16 posi i e, o al e na i e expec ed posi i e - es s (a small alue o Q25 le quan ile o - es alue dis ibu ion), i he eal - es is nega i e: W P a P a P a p es es es 1 2 2 2 3 2 (,) (, ) (, ) Simila ly: W N a N a N a n es es es 1 2 2 2 3 2 (,) (, ) (, ) S a is ics W p and W n a e chi-squa e dis ibu ed wi h deg ees o eedom equaling h ee each. Thus, he F- es has (3,3) deg ees o eedom. The es shows how co o- bo a i e - es di ec ions and alues a e ac oss h ee plan gene a ions o he pa icula TE ype. The + and -logP alues o his F- es s a is ic ac oss TE ypes a e showninFigu e4e.Thesigno helogP alue e lec s he o ing o - es s ei he up (W max =W p ;W min =W n ) o down (W max =W n ;W min =W p ) omsimula ions. Thus, long down ba s indica e signi ican cumula i e de ia ion o eal da a down om simula ions ac oss h ee plan gene a ions. The g aph (Figu e 4e) shows ha copy numbe s o all TE ypes besides Daniela in lea es o G13 we e signi i- can ly down egula ed in ela ion o he simula ions, bu his is no ue o spikes. The explana ion could be ha a high TE copy numbe in he male game es o he o i- ginal plan in e s a gene ally high simula ed le el o copy numbe s in spikes and lea es. Howe e , a plan could su i e only i he TE copy numbe is ela i ely low in diploid cells, pe mi ing, howe e , a high copy numbe in game es. Thus, plan s wi h high copy num- be s, bo h in game es and diploid cells, will be elimi- na ed om he popula ion, p o iding he obse ed di e ence be ween eal copy numbe s in game es and in diploid cells o su i o s. Mo phological and ka yo ypical changes in successi e gene a ions Ae. spel oides is a dimo phic species appea ing in na - u al popula ions as wo mo pho ypes o subspecies;Ae. spel oides ssp. ligus ica and Ae. spel oides ssp. auche i. The auche i ype is cha ac e ized by cylind ical spikes wi h widely in e spe sed spikele s. The apical spikele is awned, and he achis is ough. The ligus ica ype is cha ac e ized by dense , wo- owed ea s in which he la e al spikele s a e also awned and he achis is b i le (Figu e 1b). Signi ican mo phological and ka yo ypic abno mali ies in successi e gene a ions we e obse ed, pa icula ly: educ ion o spike awns, appea ance o in e media e ligus ica-auche i pheno ypes, abno mal quan i y o spikele s and lo e s, and spike size educ ion (Figu e 1b). Figu e 4 P inciple componen analysis (PCA).(a) PCA analysis: spikes e sus lea es o ansposable elemen (TE) copy numbe changes o e he gene a ions, no malized by he TS-84 con ol. (b) PCA analysis: wo g oups o TE, spike ac i e and lea ac i e. (c) Second p incipal componen (PC2)-based o de o TE ac i a ion. (d) No malized and cen alized pa e ns o he h ee geno ypes o Spel 52 and TEs ac oss issues and gene a ions. (e) De ia ions o eal copy numbe s in lea es and spikes om he s ochas ic model. Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 7 o 16 Fo ka yo ypic analysis we used se e al ch omosomal ma ke s in luo escen in si u hyb idiza ion (FISH) expe imen s: 5S DNA and 45S DNA p obes, ce eal cen ome e-speci ic sequence 1 (CCS-1) [28], species- speci ic Spel 1 [29] and ibe-speci ic Spel 52 [30] an- dem epea s. These p obes we e used o ch omosome iden i ica ion and localiza ion o ch omosomal ea an- gemen s. The genomic o mulas o he in es iga ed plan s we e as ollows: G9 (2n = 14 + 3B), G9 S1 o S2 (2n = 14 + 5B), G13, S1 o S3 (2n = 14), G14 (2n = 14 + 2B), G14 S1 (2n = 14 + 3B), G14 S2 (2n = 14 + 4B), and G14 S3 (2n = 14 + 3B) (Figu es 5 and 6). Dynamics o Spel 1, Spel 52, and 5S DNA clus e s in successi e gene a ions e ealed by FISH Cy ogene ical analysis e ealed d ama ic di e ences be ween genome o ganiza ions o plan s om he Kishon popula ion in compa ison wi h TS-84 plan s ep esen ing he cen e o he Ae. spel oides dis ibu ion. Fi s , Kishon plan s los a majo i y o species-speci ic andem epea Spel 1, which appea s usually as a com- ponen o e minal adenine and hymine (AT) - ich he - e och oma ic blocks in bo h a ms o almos all ch omosomes in TS-84 (Figu e 6d). Only one o wo e minal clus e s o Spel 1 pe diploid genome we e e ealed in all s udied geno ypes, which means ha up o 97% o he clus e s we e elimina ed (Figu e 5b, c, e and Figu e 6c, Table 1). We obse ed a simila pic u e o Spel 52 dis al clus e s: 57% o 82% we e elimina ed in geno ypes om he Kishon popula ion (Figu es 5 and 6, Table 1). Thus, he amoun o Spel 1 is 13 imes lowe and Spel 52 1.5 imes lowe han in he con ol geno ype TS-84 om he cen e o he species Figu e 5 Fluo escence in si u hyb idiza ion (FISH) and di e en ial s aining wi h 4’,6-diamidino-2-phenylindole (DAPI) on soma ic and meio ic ch omosomes o Aegilops spel oides (pa 1).(a) FISH wi h 5S DNA, 45S DNA and s aining wi h DAPI on soma ic ch omosomes o he o iginal G9 plan (le ). Ch omosomes 1, 6 (a ows), and B ch omosomes ca y addi ional 5S DNA si es ( igh ). (b) FISH wi h Spel 52, Spel 1 (a ows on B ch omosomes), 5S DNA and 45S DNA on he soma ic ch omosomes o he G9 S1 plan (le ); FISH wi h 5S DNA ( igh ). Ch omosomes 1 and 6 ca y addi ional 5S DNA si es (a owed). (c) F om le o igh : FISH wi h 5S DNA and 45S DNA, and DAPI on he meio ic ch omosomes o he G9 S2 plan ; 5S DNA p obe alone, ch omosomes 1, 6 (a ows), and B ch omosomes ca y addi ional 5S DNA si es; FISH wi h Spel 52 and Spel 1 on he same ch omosomes; FISH wi h CCS-1 and 45S DNA (a pe icen ic in e sion is a owed); he scheme o he main ch omosomal ea angemen s (see legend). (d) DAPI (le ) and FISH (middle) wi h 5S DNA and 45S DNA on he meio ic ch omosomes o he G13 S2 plan . A scheme o he main ch omosomal ea angemen s ( igh ). (e) FISH wi h Spel 52 and Spel 1 on he meio ic ch omosomes o he G13 S3 plan (le ). Small Spel 52 clus e (a ow) ma ks pa acen ic in e sion in he long a m o he ch omosome 5. FISH wi h 5S DNA and 45S DNA wi h DAPI s aining (middle). Bo h e mini o ch omosome 5 a e in ol ed in he e ologous synapses (whi e a ows); he e ozygous dele ion on he ch omosome 6 is shown by yellow a ow. Ch omosomes 1 and 6 ca y addi ional 5S DNA clus e s (a ows in igh ). The DNA p obes and s aining: (a-e) 5S DNA, Spel 52 and ce eal cen ome e-speci ic sequence 1 (CCS-1) in ed; 45S DNA and Spel 1 in g een; di e en ial s aining wi h DAPI in blue; (b) 5S DNA (yellow) and 45S DNA (blue) in pseudocolo s. Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 8 o 16 dis ibu ion. These da a show ha in he ma ginal Kishon popula ion a high a io o ecombina ion occu s in e minal and dis al ch omosomal egions. E iden ly, e minal clus e s o Spel 1 a e he p ima y a ge s o dele ions, ollowed by dis ally loca ed Spel 52 clus e s. To he ex en ha ch omosome e mini ha e a simila s uc u e (AT- ich he e och oma ic bands en iched wi h clus e s o Spel 1 and/o Spel 52 andem epea s and TEs), hey could be in ol ed no only in homologous bu also in he e ologous ecombina ion ollowed by ch omosomal ea angemen s. In he Kishon popula ion we obse ed an i e e sible elimina ion o e minal, spe- cies-speci ic Spel 1 andem epea s and a signi ican educ ion o he dis al, ibe-speci ic andem epea Spel 52 as a esul o high a es o homologous and he e ologous ecombina ion in his small inb ed popula ion. FISH expe imen s e ealed ha a signi ican inc ease o dec ease in copy numbe migh happen wi hou change in he clus e quan i y. Fo example, we aced a signi ican educ ion in Spel 52 copy numbe om he G9 o iginal plan o G9 S2 (Figu e 1a), while clus e numbe s inc ease om 9 o 11 in S1 and d op again o 9inS2(Table1).Inspikeso G14S29,200copieso Spel 52 a e sha ed be ween 9 clus e s while in S3 only 3,800 copies a e assigned o 10 ch omosomal clus e s (Table 1 and Addi ional ile 1). The dynamics o Spel 52 clus e s in G14 a e as ollows: he G14 o iginal line, 10 clus e s; G14 S1 o S8; G14 S2 o S9; and G14 S3 o S10 clus e s pe diploid genome. Howe e , he ch omo- somal dis ibu ion o Spel 52 clus e s signi ican ly di - e s be ween hese geno ypes. Thus, bo h homologues o ch omosome 6 ca y Spel 52 clus e s in he long a ms o he o iginal (Figu e 6a) and S1 plan s, while no clus- e s we e de ec ed ( ha is, all clus e s ha e been dele ed) as ea ly as in S1 (no shown), S2, and S3 (Fig- u e 6a and 6b). One homologue o ch omosome 5 in he G14 o iginal plan ca ies wo clus e s in he he e o- zygo e in he long a m, whe e he second clus e is a esul o ansloca ion (Figu e 6a). In S1 and S2 only one clus e s ill exis s in he he e ozygo e, which p o- ides e idence o dele ion o a second clus e as a Figu e 6 Fluo escence in si u hyb idiza ion (FISH) and di e en ial s aining wi h 4’,6-diamidino-2-phenylindole (DAPI) on soma ic and meio ic ch omosomes o Aegilops spel oides (pa 2).(a) F om le o igh : DAPI and FISH wi h he CCS-1 and 45S DNA on he meio ic ch omosomes o he o iginal G14 plan ; FISH wi h Spel 52 and Spel 1. The clus e s o Spel 1 ha ma k a homozygous pa acen ic in e sion on ch omosome 4, and a Spel 1 clus e on he B ch omosome a e a owed. FISH wi h 5S DNA and 45S DNA on he same ch omosomes. A scheme o he main ch omosomal ea angemen s. (b) FISH wi h Spel 52 and Spel 1 on he meio ic ch omosomes o he G14 S2 plan (le ). The clus e s o Spel 1 ha ma k a homozygous pa acen ic in e sion on ch omosome 4 and clus e o Spel 52 ha ma ks a he e ozygous in e sion on he ch omosome 5 a e a owed. FISH wi h 5S DNA and 45S DNA (middle). B ch omosomes ca y 5S DNA clus e s in bo h a ms. The scheme o he main ch omosomal ea angemen s ( igh ). (c) FISH wi h Spel 52 and Spel 1 on he meio ic ch omosomes o he G14 S3 plan (le ). Homozygous pa acen ic in e sion on ch omosome 4 is a owed. FISH wi h 5S DNA and 45S DNA on he same ch omosomes ( igh ). (d) Soma ic ch omosomes o TS 84, s aining wi h DAPI (le ). FISH wi h Spel 52 and Spel 1 on he same ch omosomes ( igh ). The DNA p obes and s aining: (a-d) 5S DNA, Spel 52 and ce eal cen ome e-speci ic sequence 1 (CCS-1) in ed; 45S DNA and Spel 1 in g een; di e en ial s aining wi h DAPI in blue; (a) 5S DNA (yellow) and 45S DNA (blue) in pseudocolo s. Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 9 o 16 25. Da win CD: On he o igin o species by means o na u al selec ion, o p ese a ion o a ou ed aces in he s uggle o li e. London, UK: John Mu ay 1859. 26. S ebbins GL: Adap i e adia ion o ep oduc i e cha ac e is ics in angiospe ms. 1. Pollina ion mechanisms. Ann Re Ecol Sys 1970, 1:307-326. 27. Kalenda R, Schulman AH: IRAP and REMAP o e o ansposon-based geno yping and inge p in ing. Na P o oc 2006, 1:2478-2484. 28. A agon-Alcaide L, Mille T, Schwa zache T, Reade S, Moo e G: A ce eal cen ome e sequence. Ch omosoma 1996, 105:261-268. 29. Salina EA, Pes so a EG, Ve shinin AV: Spel 1 - new amily o ce eal andem epea s. Russian J Gene 1997, 33:352-357. 30. Anam hawa -Jonsson K, Heslop-Ha ison JS: Isola ion and cha ac e iza ion o genome-speci ic DNA sequences in T i iceae species. Mol Gen Gene 1993, 240:151-158. 31. Raskina O, Ba be JC, Ne o E, Belyaye A: Repe i i e DNA and ch omosomal ea angemen s: specia ion- ela ed e en s in plan genomes. Cy ogene Gen Res 2008, 120:351-357. 32. Belyaye A, Raskina O, Ne o E: Ch omosomal dis ibu ion o e e se ansc ip ase con aining e oelemen s in wo T i iceae species. Ch omosome Res 2001, 9:129-136. 33. Mendelson D, Zoha y D: Beha iou and ansmission o supe nume a y ch omosomes in Aegilops spel oides.He edi y 1972, 29:329-339. 34. Za chi Y, Hillel J, Simchen G: Supe nume a y ch omosomes and chiasma dis ibu ion in T i icum spel oides.He edi y 1974, 33:73-180. 35. Pue as MJ: Na u e and e olu ion o B ch omosomes in plan s: a non- coding bu in o ma ion- ich pa o plan genomes. Cy ogene Gen Res 2002, 96:198-205. 36. Jones N, Houben A: B ch omosomes in plan s: escapees om he A ch omosome genome?. T ends Plan Sci 2003, 8:417-423. 37. De os KM, B own JK, Benne zen JL: Genome size educ ion h ough illegi ima e ecombina ion coun e ac s genome expansion in A abidopsis.Genome Res 2002, 12:1075-1079. 38. Sea s E: Amphidiploids in he se en-ch omosome T i icinae.Uni Missou i Ag ic Expe S a Res Bull 1941, 336:1-46. 39. Danile skaya O, Slo F, Pa lo a M, Pa due ML: S uc u e o he D osophila HeT-A ansposon: a e o ansposon-like elemen o ming elome es. Ch omosoma 1994, 103:215-224. 40. Cas o JP, Ca a e o CM: D osophila melanogas e P ansposable elemen s: mechanisms o ansposi ion and egula ion. Gene ica 2004, 121:107-118. 41. Tu ne JM, Mahade aiah SK, Fe nandez-Cape illo O, Nussenzweig A, Xu X, Deng CX, Bu goyne PS: Silencing o unsynapsed meio ic ch omosomes in he mouse. Na Gene 2005, 37:41-47. 42. Ansa i KI, Wal e S, B ennan JM, Lemmens M, Kessans S, McGahe n A, Egan D, Doohan FM: Re o ansposon and gene ac i a ion in whea in esponse o myco oxigenic and non-myco oxigenic-associa ed Fusa ium s ess. Theo Appl Gene 2007, 114:927-937. 43. Wessle SR: Tu ned on by s ess. Plan e o ansposons. Cu Biol 1996, 6:959-961. 44. Kashkush K, Feldman M, Le y AA: T ansc ip ional ac i a ion o e o ansposons al e s he exp ession o adjacen genes in whea . Na Gene 2003, 32:102-106. 45. Cheng C, Daigen M, Hi ochika H: Epigene ic egula ion o he ice e o ansposon Tos17.Mol Gene Genomics 2006, 276:378-390. 46. McClin ock B: The signi icance o esponses o he genome o challenge. Science 1984, 226:792-801. 47. Shi asu K, Schulman AH, 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 2000, 10:908-915. 48. Cha leswo h D, Cha leswo h B: T ansposable elemen s in inb eeding and ou b eeding popula ions. Gene ics 1995, 140:415-417. 49. Lewis H: The mechanism o e olu ion in he genus Cla kia.E olu ion 1953, 7:1-20. 50. Moelle DA, Gebe MA: Ecological con ex o he e olu ion o sel - pollina ion in Cla kia xan iana: popula ion size, plan communi ies, and ep oduc i e assu ance. E olu ion 2005, 59:786-799. 51. Kelly JK: Family le el inb eeding dep ession and he e olu ion o plan ma ing sys ems. New Phy ol 2005, 165:55-62. 52. K is ensen TN, So ensen P, Pede sen KS, K uho e M, Loeschcke V: Inb eeding by en i onmen al in e ac ions a ec gene exp ession in D osophila melanogas e .Gene ics 2006, 173:1329-1336. 53. G an -Down on RT, Dickinson HG: Plan s, pai ing and pheno ypes - wo’s company?. T ends Gene 2004, 20:188-195. 54. Rapp RA, Wendel JF: Epigene ics and plan e olu ion. New Phy ol 2005, 168:81-91. 55. Liu ZL, Han FP, Tan M, Shan XH, Dong YZ, Wang XZ, Fedak G, Hao S, Liu B: Ac i a ion o a ice endogenous e o ansposon Tos17 in issue cul u e is accompanied by cy osine deme hyla ion and causes he i able al e a ion in me hyla ion pa e n o lanking genomic egions. Theo Appl Gene 2004, 109:200-209. 56. Husband BC: Ch omosomal a ia ion in plan e olu ion. Am J Bo 2004, 91:621-625. 57. Oh a T: Nea -neu ali y in e olu ion o genes and gene egula ion. P oc Na l Acad Sci USA 2002, 99:16134-16137. 58. Lynch M, Cone y JS: The o igins o genome complexi y. Science 2003, 302:1401-1404. 59. Flowe s JM, Pu ugganan MD: The e olu ion o plan genomes: scaling up om a popula ion pe spec i e. Cu Opin Gene De 2008, 18:565-570. 60. REP da abase. h p://whea .pw.usda.go /ggpages/ITMI/Repea s/index.sh ml. 61. MULTILAN. h p://npsa-pbil.ibcp. /cgi-bin/npsa_au oma .pl?page=/NPSA/ npsa_mul alinan.h ml. 62. Fas PCR. h p://www.biocen e .helsinki. i/bi/p og ams/ as pc .h m. 63. Belyaye A, Raskina O: He e och oma in disc imina ion in Aegilops spel oides by simul aneous genomic in si u hyb idiza ion. Ch omosome Res 1998, 6:559-565. doi:10.1186/1759-8753-1-6 Ci e his a icle as: Belyaye e al.: T ansposable elemen s in a ma ginal plan popula ion: empo al luc ua ions p o ide new insigh s in o genome e olu ion o wild diploid whea . Mobile DNA 2010 1:6. Submi you nex manusc ip o BioMed Cen al and ake ull ad an age o : • Con enien online submission • Tho ough pee e iew • No space cons ain s o colo igu e cha ges • Immedia e publica ion on accep ance • Inclusion in PubMed, CAS, Scopus and Google Schola • Resea ch which is eely a ailable o edis ibu ion Submi you manusc ip a www.biomedcen al.com/submi Belyaye e al.Mobile DNA 2010, 1:6 h p://www.mobilednajou nal.com/con en /1/1/6 Page 16 o 16