Mitochondrial DNA and Y-chromosomal diversity in ancient populations of domestic sheep (Ovis aries) in Finland : comparison with contemporary sheep breeds

RESEARCH Open Access Mi ochond ial DNA and Y-ch omosomal di e si y in ancien popula ions o domes ic sheep (O is a ies) in Finland: compa ison wi h con empo a y sheep b eeds Ma ianna Niemi 1 , Auli Bläue 1,2 , Te hi Iso-Tou u 1 , Ve onica Nys öm 3,5 , Janne Ha jula 1,2 , Jussi-Pekka Taa i sainen 2 , Jan S o å 4 , Ke s in Lidén 4 and Juha Kan anen 1* Abs ac Backg ound: Se e al molecula and popula ion gene ic s udies ha e ocused on he na i e sheep b eeds o Finland. In his wo k, we in es iga ed hei ances al sheep popula ions om I on Age, Medie al and Pos -Medie al pe iods by sequencing a pa ial mi ochond ial DNA D-loop and he 5’-p omo e egion o he SRY gene. We compa ed he ma e nal (mi ochond ial DNA haplo ypes) and pa e nal (SNP oY1) gene ic di e si y o ancien sheep in Finland wi h mode n domes ic sheep popula ions in Eu ope and Asia o s udy empo al changes in gene ic a ia ion and a ini ies be ween ancien and mode n popula ions. Resul s: A 523-bp mi ochond ial DNA sequence was success ully ampli ied o 26 o 36 sheep ancien samples i.e. i e, se en and 14 samples ep esen a i e o I on Age, Medie al and Pos -Medie al sheep, espec i ely. Gene ic di e si y was analyzed wi hin he coho s. This ancien da ase was compa ed wi h p esen -day da a consis ing o 94 animals om 10 con empo a y Eu opean b eeds and wi h GenBank DNA sequence da a o ca y ou a haplo ype sha ing analysis. Among he 18 ancien mi ochond ial DNA haplo ypes iden i ied, 14 we e p esen in he mode n b eeds. Ancien haplo ypes we e assigned o he highly di e gen o ine haplog oups A and B, haplog oup B being he majo lineage wi hin he coho s. Only wo haplo ypes we e de ec ed in he I on Age samples, while he gene ic di e si y o he Medie al and Pos -Medie al coho s was highe . Fo h ee o he ancien DNA samples, Y-ch omosome SRY gene sequences we e ampli ied indica ing ha hey o igina ed om ams. The SRY gene o hese h ee ancien am samples con ained SNP G-oY1, which is equen in mode n no h-Eu opean sheep b eeds. Conclusions: Ou s udy did no e eal any sign o majo popula ion eplacemen o na i e sheep in Finland since he I on Age. Va ia ions in he a ailabili y o a chaeological emains may explain di e ences in gene ic di e si y es ima es and pa e ns wi hin he coho s a he han demog aphic e en s ha occu ed in he pas . Ou ancien DNA esul s i well wi h he gene ic con ex o domes ic sheep as de e mined by analyses o mode n no h-Eu opean sheep b eeds. * Co espondence: [email p o ec ed] 1 Bio echnology and Food Resea ch, MTT Ag i ood Resea ch Finland, FI-31600, Jokioinen, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle Gene ics Selec ion E olu ion © 2013 Niemi 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. Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 h p://www.gsejou nal.o g/con en /45/1/2 Backg ound A chaeological and molecula gene ic e idence sugges s ha sheep’s(O is a ies) wild ances o was he Asia ic mou lon (O. o ien alis) and ha i was domes ica ed abou 11 000 yea s ago in he Fe ile C escen egion [1]. The gene ic his o y o he domes ica ion o sheep has been in es iga ed by analysing ma e nally inhe i ed mi ochond ial DNA (m DNA) in mode n sheep b eeds. To da e, i e phylogene ically di e gen m DNA hap- log oups descending p obably om se e al O. o ien alis popula ions ha e been iden i ied in domes ic sheep [2] i.e. haplog oups A and B ha a e p esen in sheep in many pa s o he wo ld and haplog oups C, D and E ha ha e a much mo e es ic ed geog aphic ange [2,3]. Sheep sp ead ac oss Eu ope in sepa a e mig a ion epi- sodes o e ime om hei domes ica ion si e in he Nea Eas [4]. Chessa e al. [4] ha e p o ided e idence ha con empo a y sheep b eeds and popula ions o no h-wes e n and no he n Eu ope, e.g. he Finnish na- i e sheep b eeds, Finnsheep, Kainuu G ey and Åland s ill sha e gene ic ances y wi h he mos p imi i e ype o sheep in Eu ope descending om he i s immig an wa e [4]. The ea lies a cheological eco d o bu ned sheep bone in Finland da ed by adioca bon is om he la e S one Age (3 679 ± 33 BP, cal 2150–1950 BC; [5]). How- e e , such a chaeological e idence o animal husband y in Finland is sca ce and domes ica ed animals may ha e a i ed in Finland wi h he expansion o he Co ded Wa e cul u e om he sou h ia he Bal ics, bu also om he wes (Scandina ia) and eas (Russia). A ailable a chaeological da a indica e ha , in Finland, sheep hus- band y de eloped i s in ai ly limi ed a eas o he sou he n and sou h-wes e n egions. The size o he sheep popula ion was hen p obably qui e small and he a i al o any new animals e.g. wi h immig an s would ha e had a signi ican e ec on he popula ion’s gene pool. Du ing he I on Age, ag icul u e de eloped in eas - e n and cen al Finland un il eaching he no he n limi o pe manen cul i a ion (ca. 62°N la i ude) [6]. Popula ion size, demog aphy and mo phological cha ac- e s o medie al sheep in Finland can be in e ed om his- o ical ax egis e s. Fo example, in 1620, he e we e a leas 188 300 sheep in Finland [7]. Fo eign “Spanish” sheep (i.e. ances o s o mode n Me ino sheep) wi h ine wool we e impo ed in o Finland a leas om he 16 h cen u y onwa ds o se e he local wool indus y [8]. Exo ic b eeding ma e ial was in oduced mainly h ough impo ed ams, which means ha he admix u e may no ha e shaped he o iginal m DNA di e si y in he Finnish na i e sheep. Howe e , nuclea ma ke analysis o mode n sheep b eeds has shown ha impo ed animals p obably had a limi ed e ec on he gene pool o he Finnish sheep popula ion [9]. The Finnish Sheep B eeding O ganiza ion was es ablished in 1918 and b eeding o he main na i e b eed, Finnsheep, depended essen ially on animals om he eas e n pa o Finland [10]. In ecen yea s, se e al molecula and popula ion gen- e ic s udies ha e ocused on he mode n Finnish na i e sheep b eeds wi h analyses o wi hin-popula ion di e si y and popula ion s uc u e using m DNA, Y-ch omosome ma ke s, au osomal mic osa elli es and whole-genome SNP [3,9,11-15]. I has been shown ha he h ee na i e Finnish sheep b eeds, Finnsheep, Kainuu G ey and Åland and se e al o he No h-Eu opean na i e sheep b eeds belong o he No dic g oup o sho ailed sheep [13,14]. Two di e en m DNA haplog oups, A and B, seg ega e in he Finnish na i e sheep and as in o he Eu opean sheep b eeds; haplog oup B is much mo e e- quen han haplog oup A [3]. In addi ion, Meadows e al. [12] epo ed ha he h ee di e en Y-ch omosome haplo ypes assigned o wo di e en haplog oups we e p esen in Finnsheep, Kainuu G ey and Åland b eeds. In his s udy, we p esen a new app oach o he mo- lecula and popula ion gene ics o he Finnish na i e sheep b eeds based on he analysis o ancien DNA (aDNA) om ances al popula ions om he La e I on Age (800–1200 AD), Medie al (1200–1550 AD) and Pos -Medie al pe iods (1550–1800 AD). Compa ing he gene ic di e si y o ancien popula ions wi h con em- po a y b eeds can e eal empo al gene ic changes and gene pool de elopmen s, as exempli ied in a ca le Y- ch omosome s udy [16] and in a m DNA analysis o in- e ing he domes ica ion his o y o Eu opean pigs [17]. Howe e , ancien DNA s udies can also e eal ela i ely mino changes in popula ion di e si y pa e ns. Fo example, equencies o he m DNA haplog oups A and B de ec ed in ancien Chinese sheep popula ions om heB onzeAgea esimila o hoseincon empo a y Chinese sheep b eeds [18,19]. We in es iga ed he gene ic di e si y o ancien sheep emains in Finland by sequencing a 523-bp m DNA D- loop sequence and a 130-bp segmen in he 5’-p omo e egion o he o ine sex de e mining egion Y (SRY) gene o de ec a biallelic Y-ch omosome SNP ma ke oY1. We compa ed m DNA and Y-ch omosome ma ke da a om ancien and mode n domes ic sheep popula ions in Finland and o he pa s o Eu ope and Asia and s udied empo al changes in ma e nal and pa e nal gene ic di- e si y and m DNA di e si y pa e ns. Fo he m DNA s udy, wo mode n da ase s we e used: (1) 10 mode n Eu opean sheep b eeds we e sequenced and (2) add- i ional m DNA sequences we e collec ed om GenBank ha spanned he same m DNA D-loop nucleo ide si es as hose p esen in he ancien m DNA sequences. To ou knowledge, his is he i s ime ha m DNA and Y- ch omosome gene ic di e si ies o ancien sheep popula- ions a e in es iga ed in he same s udy. Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 2 o 14 h p://www.gsejou nal.o g/con en /45/1/2 Ma e ials and me hods Ancien sheep bone ma e ial Fo he aDNA analysis, 36 sheep bones we e selec ed om 14 di e en loca ions and 18 di e en a chaeological exca- a ions ac oss Finland and one si e in no he n No way (Table 1, Addi ional ile 1: Figu e S1). As a ule, unbu ned bone om ime pe iods p io o he La e I on Age is no p ese ed in he Finnish acid soil; hus ea lie bone ma e - ial consis s o small bu ned agmen s ha a e no use ul o aDNA analyses [20]. The ea lies unbu ned bones a ailable o his s udy de i e om he La e I on Age (800–1200 AD). When possible, me aca pal o me a a sal bones we e selec ed o wo easons: (1) hese bones a e easy o dis inguish be ween sheep and goa s and (2) hey a e o en ound in a comple e o semi-comple e s a e in he exca a ions because since hey a e nu i ionally poo hey we e no usually bu che ed o cooking. Howe e , o he bone elemen s we e also sampled, e.g. when me apodials we e no p esen in he sample, o we e badly p ese ed o agmen ed, o i hey did no o m he highes Minimum Numbe o Indi iduals (MNI) wi hin he exca a ion o phase. In some cases, only bones ha a e di ec ly iden i iable as om sheep o goa (e.g. ee h) we e selec ed because i is possible o iden i y a species om analyses o he m DNA D- loop egion. Ca e was aken no o sample he same indi idual wice i.e. wi hin one si e and phase, elemen s loca ed on he same side we e selec ed o a gi en sample, o he wise age and size o he animal we e used o sepa a e di e en indi iduals. All samples ha based on he a chaeological con ex we e assumed o o igin- a e om he I on Age o Medie al pe iod we e adio- ca bon-da ed. In addi ion, Pos -Medie al samples o which da ing was unce ain om he a chaeological con ex , we e also adioca bon-da ed. Nine een sheep bones o ee h we e adioca bon-da ed in he Labo a- o y o Ch onology o he Finnish Museum o Na u al His o y, Uni e si y o Helsinki (Table 1). I on age Ele en samples exca a ed a ou di e en si es om he I on Age we e included in his s udy (Table 1, Addi ional ile 1: Figu e S1). The aDNA samples (sample ID: OaM1-5) belong o he Viking Age phase o he Mulli si e (1090–930 BP, Table 1). The animal bone assemblage om his si e is mainly composed o domes ic animals, bu also con ains a a ie y o wild mammals. Eas e n Finland Viking Age si es we e sampled (OaSys1 and OaMik1) and samples om wo I on Age ceme e y si es in sou h-wes and sou he n Finland (OaLui1-2 and OaKi 1) we e collec ed. The sample om B od ko bne- se , Pas ik (No way) (OaPas1) de i es om a ec angula hea h o Sami o igin and also da es o he la e I on Age (984 ± 31 BP, Table 1). Medie al pe iod Ele en samples om wes e n and eas e n Finland om he Medie al pe iod we e analysed (Table 1, Addi ional ile 1: Figu e S1) among which eigh o igina ed om he own o Tu ku (OaÅA3-6, OaVe 1-4), he la ges Medie al own in Finland ha is si ua ed on he sou h- wes coas by he i e Au a, one om B idge ine Abbey o Naan ali (OaNaa1) si ua ed in sou h-wes Finland ca. 18 km om Tu ku, one om Köka monas- e y, a medie al F anciscan monas e y si ua ed in he Åland a chipelago (OaKök1), and one om eas e n Finnish Ka elia i.e. a dese ed G eek O hodox illage in Papinniemi in Uukuniemi (OaUuk2). Pos -Medie al pe iod Fou een samples om he Pos -Medie al pe iod we e analysed (Table 1, Addi ional ile 1: Figu e S1) among which six we e collec ed om he sho es o he Gul o Bo hnia, om Pie a saa i (OaPie1), To nio (OaTo 1-2) and Oulu (OaOul1-3), se en om sou h-wes and sou h- e n Finland i.e. wo om Köka monas e y (OaKök2-4), wo om Tu ku (OaÅA1-2) and wo om Helsinki (OaHel1-2), bo h in u ban con ex s and one om an in- land egion a Pih ipudas (OaPih1). Mode n sheep The occu ence o ancien sheep m DNA sequences in mode n sheep b eeds was in es iga ed wi h wo da a- se s: (1) ou own da ase o 94 un ela ed animals om 10 sheep b eeds o local a ie ies [see Addi ional ile 1: Figu eS1]and(2)ada ase composedo GenBankse- quence da a o 50 Eu opean and Asian sheep b eeds p e iously published. Ou own da ase was also used o he compa ison be ween ancien and mode n DNA. P e iously desc ibed in [3,14], b eeds included in ou own da ase a e he ollowing: he Finnish na i e sheep b eeds Finnsheep, Kainuu G ey and Åland, he Viena sheep om Russian Ka elia, Bozakh sheep om he Caucasus, Romano and Opa ino sheep om cen al Russia, Olkuska sheep om Poland, he mo phological ype o Vlashko Vi o oga–P amenka sheep om Se bia and Ox o d Down sheep om he UK. All sheep b eeds we e locally de eloped b eeds excep o Ox o d Down, which is a syn he ic comme cial English b eed [40]. Kainuu G ey, Åland, Opa ino, Olkuska and Vlashko Vi o oga-P amenka sheep ha e unde gone a educ ion in popula ion size du ing he las 10 yea s and a e a e o endange ed b eeds. Resul s om a Y-ch omosome SNP oY1 analysis in ancien sheep we e compa ed wi h hose o a global sheep Y-ch omosome s udy including Finnsheep, Kainuu G ey and Åland sheep [12]. Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 3 o 14 h p://www.gsejou nal.o g/con en /45/1/2 DNA ex ac ion and labo a o y me hods Labo a o y en i onmen and DNA ex ac ion o ancien sheep samples Bone samples we e p epa ed by emo ing he ou e laye o he bones and collec ing 50 o 200 mg o bone powde wi h a d ill. DNA ex ac ion was ca ied ou in an ai -con olled s e ile labo a o y and in a lamina low hood (EU-14 HEPA il e ed ai unde posi i e ai p es- su e isola ion). Sepa a e labo a o ies we e used o sam- ple p epa a ion, DNA ex ac ion and PCR ampli ica ion. Table 1 Summa y o ancien samples s udied in his a icle Sample Da ing aDNA Sample ID Loca ion Si e Bone ype *BP (± 1σ) **Pe iod m DNA SRY gene OaMik1 Mikkeli Moisio La oka ano [21] Mandible 865 ± 33 I on Age No OaM1 Raisio Mulli [22] Me aca pal 965 ± 30 I on Age Yes un eadable OaM2 Raisio Mulli [22] Me aca pal 1040 ± 31 I on Age Yes No OaM3 Raisio Mulli [22] Me aca pal 955 ± 30 I on Age Yes un eadable OaM4 Raisio Mulli [22] Me aca pal 995 ± 30 I on Age Yes un eadable OaM5 Raisio Mulli [22] Me aca pal 1081 ± 32 I on Age Yes un eadable OaSys1 Sysmä Ihananiemi [23] Too h (Mola ) 1093 ± 31 I on Age No OaKi 1 Tu ku Ki kkomäki [24] Too h agmen s NA I on Age No OaLui1 Eu a Luis a i [25,26] Too h agmen s NA I on Age No OaLui2 Eu a Luis a i [25,26] Too h agmen s NA I on Age No OaPas1 Pas ik (No way) B od ko bnese [27] Me a a sal 984 ± 31 I on Age Pa ial No OaNaa1 Naan ali Luos a i [28] Me aca pal 452 ± 30 Medie al Yes Yes OaUuk2 Uukuniemi Papinniemi [29] Me a a sal 410 ± 30 Medie al Yes No OaÅA3 Tu ku Åbo Akademi [30] Me aca pal 506 ± 32 Medie al Yes No OaÅA4 Tu ku Åbo Akademi [30] Me aca pal 581 ± 31 Medie al Yes No OaÅA5 Tu ku Åbo Akademi [30] Me aca pal 722 ± 32 Medie al Yes No OaÅA6 Tu ku Åbo Akademi [30] Me aca pal 737 ± 32 Medie al Yes No OaVe 1 Tu ku Aboa Ve us [28,31] Ho nco e 487 ± 30 Medie al No OaVe 2 Tu ku Aboa Ve us [28,31] Maxilla DBC Medie al No OaVe 3 Tu ku Aboa Ve us [28,31] Axis 550 ± 30 Medie al No OaVe 4 Tu ku Aboa Ve us [28,31] Lowe jaw DBC Medie al No OaKök1 Köka Klos e [32] Me aca pal 489 ± 30 Medie al Yes No OaHel1 Helsinki Snellmaninka u [33] Me a a sal DBC Pos -Med Yes No OaHel2 Helsinki Snellmaninka u [33] Me a a sal DBC Pos -Med Yes No OaOul1 Oulu Kajaaninka u [34] Radius DBC Pos -Med Yes No OaOul2 Oulu Lyseo [35] Radius DBC Pos -Med Yes No OaOul3 Oulu Pikisaa i [36] Radius DBC Pos -Med Yes No OaPie1 Pie a saa i Lass olk [37] Me aca pal DBC Pos -Med Yes No OaTo 1 To nio Keskika u [38] Me aca pal DBC Pos -Med Yes Yes OaTo 2 To nio Keskika u [38] Me aca pal DBC Pos -Med Yes No OaÅA1 Tu ku Åbo Akademi [30] Me aca pal DBC Pos -Med Yes un eadable OaÅA2 Tu ku Åbo Akademi [30] Me aca pal DBC Pos -Med Yes No OaPih1 Pih ipudas Hämeensaa i [39] Tibia 342 ± 30 Pos -Med Yes No OaKök2 Köka Klos e [32] Me aca pal 305 ± 30 Pos -Med Yes Yes OaKök3 Köka Klos e [32] Me aca pal DBC Pos -Med Yes No OaKök4 Köka Klos e [32] Me aca pal DBC Pos -Med Yes No Iden i y o samples include: sample ID, loca ion i.e. own whe e samples we e exca a ed, a chaeological si e, and ype o bone (museums ID a ailable upon eques ); da ing o samples include adioca bon-da es (*BP (± 1σ)) o da es es ima ed by he con ex (DBC) and co esponding his o ical **pe iods (Pos -Med = Pos -Medie al); esul s o m DNA D-loop and Y-ch omosomal SRY gene sequence analysis a e indica ed o each sample; only he samples yielding m DNA we e analysed o he SRY gene. Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 4 o 14 h p://www.gsejou nal.o g/con en /45/1/2 The labo a o ies and equipmen we e UV- ea ed and bleach was used o clean he lamina and labo a o y egu- la ly. P o ec i e whole body sui s, double glo es and masks we e used inside he aDNA labo a o y. Two aDNA labo a- o ies pa icipa ed in aDNA analyses (MTT Ag i ood Re- sea ch Finland and S ockholm Uni e si y, Sweden). DNA ex ac ion was ca ied ou as desc ibed in [41]. The bone powde was diges ed in 900 μL 0.5 M EDTA, 100 μL 10 M u ea and 5 μL p o einase K (20 mg/ml) wi h cons an s i ing a 55°C o e nigh . A e cen i u- ga ion (2000 pm o 5 min), he supe na an was con- cen a ed and DNA was ex ac ed wi h a QIAquick PCR Pu i ica ion Ki (Qiagen, Sweden) acco ding o he man- u ac u e ’s ins uc ions. DNA ma ke s, p ime design and PCR A 523-bp sequence o m DNA encompassing pa o he m DNA D-loop om ancien sheep ([GenBank: NC001941] posi ions 15,978-16,501) was analysed. A 130-bp sequence in he 5’-p omo e egion o he SRY gene on o ine Y-ch omosome ([GenBank:AY604734] posi ions 58–179) was analysed o de ec he Y- ch omosome SNP ma ke (oY1) [GenBank:AY604734.2: g.67, A>G]. P ime s [see Addi ional ile 2: Table S1] o i e o e - lapping agmen s o m DNA D-loop sequence and one Y-ch omosome SRY gene agmen we e designed wi h P ime 3 [42] using m DNA and SRY gene e e - ence sequences [GenBank:NC001941] and [GenBank: AF026566.1], espec i ely. PCR o ancien DNA was pe o med in 25 μL mix u e ha con ained 1× PCR bu e (Qiagen, Sweden), 0.2 μM o each p ime , 0.4 mM dNTP, 2.5 mM MgCl 2 (Qiagen, Sweden), 0.25 uni s (U) o U acil DNA Glycosylase (UNG, Sigma-Ald ich), 1.5 U o Ho S a Taq DNA Poly- me ase (Qiagen, Sweden) and 5–10 μL o DNA ex ac . The PCR p og am consis ed in ini ial s eps a 37°C o 10 min and 95°C o 15 min ollowed by 55 h ee-s ep cycles a 94°C o 30 s, AT°C o 40s and 72°C o 1 min and a inal s ep a 72°C o 10 min. AT s ands o he annealing empe a u e speci ic o each p ime se [see Addi ional ile 2: Table S1]. Au hen ici y o ancien sheep DNA Common measu es o p e en con amina ion we e used [43,44], such as sepa a ed a eas o sample p epa a ion, ancien DNA analyses and p e-PCR, wea ing p o ec i e clo hing, using disposable ools and pipe es wi h ae osol esis an il e ips and ea ing equipmen and wo king su aces wi h bleach and ul a- iole i adia ion e- quen ly (see DNA ex ac ion and labo a o y me hods). O e lapping p ime s speci ic o sheep DNA we e designed o p e en possible annealing o human DNA and we e checked by ampli ica ion (see DNA ex ac ion and labo a o y me hods). The au hen ici y o he aDNA analysis was con olled a a ious s eps o he labo a o y wo k- low. In gene al, when consis en sequences we e ob ained in h ee o mo e ampli ica ions, he sequence was conside ed ep o- ducible and accep ed as au hen ic. In he aDNA ex ac- ion s ep, 13 o he samples (ou o he 26 samples which we e success ully ampli ied o he 523-bp m DNA D- loop sequence) (Table 1) we e ex ac ed se e al imes, and a leas wo PCR- eac ions we e pe o med o sepa a e DNA ex ac ions o ampli y and sequence each o e lap- ping m DNA D-loop agmen a leas wice. Applying hese s ic ex ac ion and ampli ica ion s eps aimed a con i ming he ep oducibili y o ou aDNA sequence p o ocol and no sequence anomalies we e de ec ed. Mo e- o e , i e samples (OaM1, OaNaa1, OaTo 2, OaÅA5 and OaÅA6) we e ex ac ed, ampli ied and sequenced in wo di e en aDNA labo a o ies (MTT Ag i ood Resea ch Finland and S ockholm Uni e si y, Sweden) and he esul s we e iden ical. A each s ep o he aDNA ex ac- ion and ampli ica ion p ocedu e, nega i e con ols we e pe o med. In he i s ex ac ion o ou aDNA ial, a mammo h sample was used as posi i e con ol, which cons i u es a sui able con ol because a p e ious s udy [45] has shown ha i does con ain mammo h DNA, i does no con ain mode n DNA and i s sequence clea ly di e s om ha o sheep. Mi ochond ial DNA ampli ica ion o mode n sheep samples DNA om 94 mode n sheep o 10 Eu opean and Asian b eeds [3,14] we e analysed. Addi ional ile 2: Table S1 includes p ime s o he ampli ica ion o a 664-bp se- quence o m DNA [see Addi ional ile 2: Table S1]. The same PCR eac ion mix was used o hese samples han o he aDNA samples excep ha UNG was no included. PCR condi ions we e as ollows: 95°C o 15 min ollowed by 32 h ee-s ep cycles a 94°C o 30 s, 58°C o 40 s and 72° o 1 min and a inal s ep a 72°C o 10 min. 500 ng o empla e DNA we e used pe eac ion. Y-ch omosome ma ke da a we e a ailable om a p e ious s udy [12]. Sequencing o PCR p oduc s PCR p oduc s o mode n and ancien DNA samples we e pu i ied using ExoSAP-IT enzyme (GE Heal hca e Li e sciences, UK). Sequencing eac ions we e pe o med using DYEnamic ET Te mina o Ki (GE Heal hca e Li e sciences, UK). The sequencing p oduc s we e pu i ied by e hanol p ecipi a ion and sepa a ed on MegaBACE1000 ™ (Ame sham Biosciences, UK). Bo h s ands o each ag- men we e sequenced and he same p ime s we e used o bo h sequencing and agmen ampli ica ion. Se- quence da a we e base-called wi h Cima on 3.12 Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 5 o 14 h p://www.gsejou nal.o g/con en /45/1/2 basecalle using a MegaBACE Sequence analyze . 3.0.0111.1603 (Ame sham Biosciences, UK). Sequences we e analysed wi h Sequenche 4.6 8 (Gene Codes, Ann A bo , MI). S a is ical analyses Mi ochond ial DNA sequences om 94 mode n sheep [GenBank: JX484017-JX484110] and 26 ancien sheep [GenBank: JX484111-JX484137] we e aligned using CLUS- TALW2 [46] (penal ies o gap opening, gap ex ension, gap dis ances we e 10, 0.20 and 5, espec i ely). When s a is- ical analyses we e pe o med o a subse o sequences, he CLUSTALW2 alignmen was ca ied ou sepa a ely o each coho . The size o he aligned m DNA sequence was 523-bp. One I on Age sample (OaPas1), o which only 297-bp o he m DNA D-loop was sequenced, was omi ed om he s a is ical analyses (Table 1). A co esponding CLUSTALW2 alignmen was done o he Y-ch omosome 130-bp sequences [JX484138- JX484140]. Alignmen gaps we e excluded om he s a is ical analyses. The app op ia e DNA subs i u ion model o analyse ou m DNA da a was he Hasegawa-Kishino-Yano model ([47]; HKY85)+Гsuppo ed bo h by FindModel [48] web se e (h p://www.hi .lanl.go /con en /sequence/ indmodel/ indmodel.h ml) and MEGA5.05 p og am [49] and selec ed on he basis o he Akaike in o ma ion c i e - ion (AIC) [48] o di e en models. The phylogene ic ana- lysis o he 94 mode n and 26 ancien sheep sequences and one ou g oup sequence om u ial sheep (O is ignei bocha iensis [GenBank:AF039580.1]) was conduc ed using wo app oaches. MEGA5.05 was used o cons uc he Neighbo -joining (NJ) ee wi h 1000 boo s ap eplica es. Howe e , he Tamu a-Nei subs i u ion model [50] wi h a Г dis ibu ion pa ame e alue α= 0.05 was used in he ana- lysis because HKY is no implemen ed in he MEGA so - wa e and he Tamu a-Nei model was suppo ed by AIC. The maximum likelihood (ML) analysis was pe o med using PhyML . 3.0 p og am [51] and he HKY85+Г(lnL = −1271.48434, Гdis ibu ion pa ame e alue α= 0.047). Boo s ap suppo o a ML ee was calcula ed using 1000 boo s ap eplica es and he ee was d awn wi h he T ee- View p og am . [52]. In addi ion, a median-joining ne - wo k (wi h ε= 0 o be mos conse a i e) be ween he haplo ypes was cons uc ed and misma ch dis ibu ion was pe o med using NETWORK 4.6.0.0 [53]. The ollowing pa ame e s we e calcula ed o es ima e he gene ic di e si y o he m DNA da a in he di e en coho s (i.e., in he ancien popula ions and mode n b eeds): numbe o haplo ypes (h), numbe o seg ega - ing si es (S), haplo ype di e si y (Hd = p obabili y ha wo m DNA sequences chosen andomly om he sam- ple a e di e en ), nucleo ide di e si y (π= numbe o nu- cleo ide di e ences be ween andomly chosen pai s o sequences), and a e age numbe o nucleo ide di e ences (K) (DnaSP .5 [54]). The a-es ima es (‘ he a’θ=N e μ, whe e N e is he e ec i e popula ion size in he case o a haploid locus and μis he o e all mu a ion a e a he haplo ype le el) we e compu ed using ARLEQUIN . 3.5 [55]: he expec ed le el o di e si y, θ S, was de i ed om he obse ed numbe o seg ega ing si es S and θ π om he obse ed mean numbe o pai wise nucleo ide di e - ences π. Fu he mo e, Tajima’s D es s a is ic was com- pu ed using DnaSP. We in es iga ed he dis ibu ion o he ancien m DNA sequences p esen among he 94 sequences o he mode n sheep samples sequenced he e and sea ched o sha ed iden ical sequences in he GenBank DNA da abase (NCBI/BLAST, h p://blas .ncbi.nlm.nih.go / Blas .cgi). The haplo ypes we e de e mined using DnaSP [see Addi ional ile 3: Table S2]. The ancien Y-ch omosome SNP oY1 da a we e com- pa ed wi h mode n da a [9,12] and geog aphic equency dis ibu ions o oY1 alleles. The mode n Y-ch omosome SNP da a [9,12] on he equencies o oY1 alleles a e p esen ed in Addi ional ile 4: Table S3. Resul s Radioca bon da ing Nine een samples we e success ully adioca bon-da ed, while o h ee samples, he quan i y o collagen in he enamel was no su icien o his echnique (Table 1). The adioca bon da ing indica ed ha wo samples, (OaPih1, OaKök2), we e younge and one (OaUuk2) was olde han in e ed om he a chaeological eco ds om he same si es (Table 1). The da ing esul s indica ed ha he oldes samples success ully analyzed o m DNA we e app oxima ely 1000 yea s old. Success a e o aDNA analyses Mi ochond ial DNA analysis was ca ied ou on 36 an- cien sheep samples and m DNA ampli ica ion was success ul o 27 samples, including he pa ial m DNA sequence o he OaPas1 sample (Table 1). Mos o hese 27 samples equi ed se e al ampli ica ions pe agmen in o de o ob ain a leas h ee good-quali y sequences om a leas wo PCR eac ions. Excluding he samples o which no m DNA ampli ica ion was ob ained and conside ing all sepa a e ampli ica ions o he i e m DNA D-loop agmen s, he ampli ica ion success a e was 46%, 56% and 68% o sheep samples om he I on Age, Medie al and Pos -Medie al pe iods, espec i ely. A e age ampli ica ion success a es and agmen leng hs o he m DNA D-loop sequence a e summa ised in [Addi ional ile 2: Table S1]. As expec ed, wi h aDNA samples, he highes success a e was ob ained when he ampli ied ag- men s o m DNA D-loop we e sho es . In addi ion, among he aDNA samples om di e en pe iods, hose om he Pos -Medie al pe iod had he highes success Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 6 o 14 h p://www.gsejou nal.o g/con en /45/1/2 a e, which is explained by he ac ha Pos -Medie al ma- e ial is abundan allowing a mo e c i ical p io -selec ion o bones han o olde ma e ials. Rep oducible sequences we e ob ained o se en o he 11 Medie al samples and o i e (plus one pa ial) o he 11 I on Age samples (Table 1). Ampli ica ion o a 130-bp sequence o Y-ch omosome 5’p omo e egion in he SRY gene was es ed in aDNA samples ha we e success ully ampli ied o m DNA and h ee con ained he SRY gene sequence. As expec ed, among he di e en aDNA analyses, ampli ica ion suc- cess a e was lowes (12%) o he agmen con aining he SRY gene. Mi ochond ial DNA haplo ypes Alignmen be ween he 94 mode n and 26 ancien DNA sequences (OaPas1 con aining a pa ial m DNA se- quence was excluded) e ealed 47 SNP (46 ansi ions and one ans e sion) and h ee inse ion-dele ions [see Addi ional ile 5: Figu e S2]. When conside ing he alignmen o he aDNA samples only, 27 ansi ions and no ans e sion we e ound. Fi y-six haplo ypes (18 in he ancien and 46 in he mode n DNA sequences) we e iden i ied among he 121 mode n and ancien sheep samples (Figu e 1, Tables 2 and 3). Mo e han hal o hese haplo ypes (30 o 56) we e p i a e o one indi id- ual and hus o one mode n sheep b eed o ancien sheep popula ion indica ing high a ia ion in he ana- lysed D-loop egion. The NJ and ML analyses ga e simila phylogene ic opologies hus only he ML ee is p esen ed in Figu e 1. Two highly di e gen domes ic sheep lineages, o ine m DNA haplog oups A and B, we e de ec ed wi h ela- i ely high s a is ical suppo . As expec ed, compa isons wi h e e ence sequences [GenBank: AF039577.1] and [GenBank: AF039578.1], haplog oups B and A we e e- spec i ely he majo and mino haplog oups. F equen- cies o haplog oups A and B we e 0% and 100% o he I on Age, 28.6% and 71.4% o he Medie al and 14.3% and 85.7% o he Pos -Medie al sheep coho s, espec - i ely and 21.9% and 78.1% o he Finnish mode n sheep b eeds. In addi ion, he Ne wo k analysis and misma ch dis ibu ion suppo ed he exis ence o hese wo di e - gen haplog oups in he mode n sheep b eed da a p e- sen ed in Figu es S3 and S4 [see Addi ional ile 6: Figu es S3 and S4]. These analyses ep oduced he well- es ablished obse a ions wi h a s a -shaped pa e n in he o ine m DNA haplo ype ne wo k and a smoo h shape o misma ch dis ibu ion indica ing a popula ion expansion in he his o y o he species. Simila pa e ns we e obse ed in he Medie al-I on Age and Pos -Medie al coho s when analysed sepa a ely ( esul s no shown). We analysed he dis ibu ion o he 18 ancien Finnish sheep m DNA haplo ypes in he mode n sheep b eeds by sequencing samples o 10 Eu asian sheep b eeds and by sea ching o sha ed sequences in he GenBank DNA da abase. The sha ed haplo ype analysis showed ha 14 o he 18 ancien haplo ypes we e p esen in he mod- e n sheep b eeds and in he GenBank DNA da abase (Figu e 1) and [Addi ional ile 3: Table S2]. In he mod- e n sheep popula ions, equencies o ancien haplo- ypes we e highes in he na i e Finnish sheep b eeds, Finnsheep (0.38) and Kainuu G ey (0.53) wi h six an- cien haplo ypes i.e. H11 in haplog oup A and H22, H32, H40, H42, H44 in haplog oup B. The wo ancien haplo ypes, H05 in haplog oup A and H32 in hap- log oup B a e common haplo ypes while o he a e ancien haplo ypes exis e.g. in con empo a y na i e b eeds om he Caucasus, Russia and Se bia [see Addi ional ile 3: Table S2]. The ou ancien haplo ypes H06, H30, H43 and H50 we e absen bo h in ou own mode n da ase and in he GenBank DNA da abase [see Addi ional ile 3: Table S2]. Analysis o popula ion di e si y The s a is ics summa ising he le el o m DNA a ia ion in he Finnish ancien popula ions and mode n sheep b eeds and in se en o he mode n Eu asian b eeds a e p esen ed in Tables 2 and 3, espec i ely. Addi ional ile 1: Figu e S1 shows he si es om which he Finnish a chaeo- logical samples we e collec ed and o which m DNA was success ully ampli ied. The I on Age coho comp ising only i e samples and exca a ed a a single a chaeological si e on he p ehis o ical a m Mulli displayed he lowes alues o all di e si y es ima es (s, h, Hd, K, and π). Haplo ype di e si y was highe in ancien sheep samples om he Medie al and Pos -Medie al pe iods (Hd = 1.0 and 0.98, espec i ely) han in any o he mode n popula- ions. The Pos -Medie al samples o igina ed om se e al a chaeological si es, while he Medie al samples we e exca a ed mainly in Tu ku. Among he mode n sheep b eeds analysed, Bozakh, Finnsheep and P amenka showed he highes haplo ype di e si y (Hd > 0.90; Tables 2 and 3) while Olkuska and Åland had he lowes haplo ype di e si y (Hd = 0.75; Tables 2 and 3). The Finnsheep sam- ples we e collec ed in se e al sheep locks om di e en pa s o Finland. The measu es aking in accoun he mo- lecula na u e o he da a showed highes a ia ion in he popula ions and b eeds in which bo h haplog oups A and B we e seg ega ing. Nucleo ide di e si y a ied among coho s and was highes (π=15.51 * 10 -3 ) in P amenka sheep, o igina ing om he Balkan egion, which was one o he main dispe sal ou es o Nea Eas e n domes i- ca ed sheep en e ing Eu ope. The di e si y es ima e θs is in luenced by gene ic bo le- necks, whe eas θπ is ela i ely insensi i e. Consequen ly, θπ-θs will be nega i e in s able popula ions unde an in ini e-si es model o mu a ion-d i equilib ium and Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 7 o 14 h p://www.gsejou nal.o g/con en /45/1/2 Figu e 1 Maximum Likelihood ee o m DNA haplo ype sequences ound wi hin mode n (black) and ancien ( ed) sheep. B anch opog aphy suppo ed by boo s ap alues g ea e han 50% is indica ed; he numbe o samples sha ing haplo ypes is gi en in b acke s; b eed names a e abb e ia ed as Ox o d Down (Ox ), Olkuska (OL), P amenka (P a), Bozakh (Boz), Opa ino (Opa), Romano (Ro), Viena (Vie), Kainuu G ey Sheep (Kai), Åland (ALD) and Finnsheep (Fin); ancien Finnish sheep samples a e di ided in o h ee pe iods: I on Age (I on), Medie al (Med) and Pos -Medie al (Pos ) acco ding o adioca bon o con ex da ing (see Table 1); he ee is oo ed wi h a sequence om u ial sheep (O is ignei bocha iensis) [GenBank: AF039580.1]. Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 8 o 14 h p://www.gsejou nal.o g/con en /45/1/2 posi i e as a esul o a educ ion in he numbe o seg e- ga ing si es. In ou da a, θπ-θs es ima es we e nega i e i.e. -1.465 and −1.563 o Pos Medie al and mode n Finnsheep, espec i ely (Table 2), whe eas hey we e posi- i e o Åland, Kainuu G ey, Olkuska, and P amenka sheep, and sligh ly posi i e o he P amenka b eed and he Medie al popula ion (Tables 2 and 3). S a is ically signi i- can Tajima’sD- alueswe eposi i e o KainuuG eyand Olkuska sheep, indica ing a educ ion o m DNA di e si y. Y-ch omosome analysis Fo h ee o he 27 ancien sheep samples, Y-ch omosome SRY sequences we e ampli ied and we e hus gene ically iden i ied as ams (Table 1) wi h one am sample om he Medie al pe iod and wo om he Pos -Medie al pe iod (Table 1). All h ee ancien Finnish sheep had SNP G-oY1 in he SRY gene. In he mode n Finnish b eeds, he equency o G-oY1 was 77%, 57%, and 60% in Finnsheep, Kainuu G ey, and Åland, espec i ely [see Addi ional ile 4: Table S3], indica ing ha SNP G-oY1 is equen in he mode n no h-Eu opean b eeds (Figu e 2). Discussion We success ully analysed m DNA and Y-ch omosome di e si y in ancien sheep emains om Finland and ound ha Finnish sheep gene ic di e si y has been qui e cons an o e he las 1000 yea s. Ou aDNA esul s i well wi h he gene ic con ex o he mode n no h- Eu opean domes ic sheep b eeds analysed ei he p e i- ously [3,12] o in he p esen s udy. Bo h o ine hap- log oups A and B ha e been p esen in he Finnish sheep popula ion o a leas mo e han 700 yea s and no ema kable empo al changes in hei equencies ha e occu ed. Fou o he 26 ancien m DNA sequences Table 3 Summa y s a is ics o mode n Polish, Russian and UK b eeds S a is ics Mode n sheep b eeds Bozakh Olkuska Opa ino Ox o d Down P amenka Romano Viena N9999989 S 21 14 16 9 19 16 11 h8455756 Hd 0.97 0.75 0.86 0.81 0.94 0.89 0.89 K 7.44 7.39 5.89 2.83 8.11 6.96 4.39 π14.23 14.13 11.26 5.42 15.51 13.32 8.39 D−0.180 2.085* 0.002 −0.664 0.786 0.661 0.397 θs 7.727 5.151 5.887 3.311 6.991 6.171 4.047 θπ 7.667 7.389 5.889 2.833 8.333 6.964 4.389 θπ-θs−0.06 2.238 0.002 −0.478 1.342 0.793 0.342 N, numbe o indi iduals sampled; S, numbe o seg ega ing si es (excluding indels); h, numbe o haplo ypes; Hd, haplo ype di e si y; K, a e age numbe o di e ences; π, nucleo ide di e si y*10 -3 ; D, Tajima´s D s a is ic alue whe e s a is ically signi icances P < 0.05 a e ma ked wi h *; θs, ‘The a’de i ed om he obse ed numbe o seg ega ing si es S and θ π om he obse ed mean numbe o pai wise nucleo ide di e ences π. Table 2 Summa y s a is ics o ancien Finnish sheep popula ions om h ee pe iods and mode n Finnish sheep b eeds S a is ics Ancien Finnish sheep Mode n Finnish sheep b eeds Pos -Medie al Medie al I on Age Åland Finnsheep Kainuu G ey N 14 7 5 9 15 8 S 25 18 2 17 14 15 h 12 7 2 4 10 4 Hd 0.98 1.00 0.40 0.75 0.93 0.82 K 6.13 7.81 0.80 7.94 2.61 8.11 π11.72 14.93 1.53 15.19 4.99 15.50 D−0.934 0.352 −0.973 1.315 −1.562 2.054* θs 7.861 7.347 0.960 6.255 4.306 5.785 θπ 6.396 8.095 0.800 7.944 2.743 8.107 θπ-θs−1.465 0.748 −0.16 1.689 −1.563 2.322 N, numbe o sampled indi iduals; S, numbe o seg ega ing si es (excluding indels); h, numbe o haplo ypes; Hd, haplo ype di e si y; K, a e age numbe o di e ences; π, nucleo ide di e si y*10 -3 ; D, Tajima´s D s a is ic alue whe e s a is ically signi icances P < 0.05 a e ma ked wi h *; θs, ‘The a’de i ed om he obse ed numbe o seg ega ing si es (S) and θ π om he obse ed mean numbe o pai wise nucleo ide di e ences (π). Niemi e al. Gene ics Selec ion E olu ion 2013, 45:2 Page 9 o 14 h p://www.gsejou nal.o g/con en /45/1/2