Ectopic kit copy number variation underlies impaired migration of primordial germ cells associated with gonadal hypoplasia in cattle

Ec opic KIT Copy Numbe Va ia ion Unde lies Impai ed Mig a ion o P imo dial Ge m Cells Associa ed wi h Gonadal Hypoplasia in Ca le (Bos au us) Heli Venho an a1*☯, Hube Pausch2☯, Michal Wysocki2, Izabela Szcze bal3, Ree a Hänninen4, Juhani Taponen1, Pekka Uima i5, K zysz o Flisikowski6, Hannes Lohi4, Ruedi F ies2, Ma ek Swi onski3, Magnus Ande sson1 1 Depa men o P oduc ion Animal Medicine, Uni e si y o Helsinki, Saa en aus, Finland, 2 Chai o Animal B eeding, Technische Uni e si ä München, F eising-Weihens ephan, Ge many, 3 Depa men o Gene icsandAnimal B eeding, Uni e si y o Li e Sciences, Poznań, Poland, 4 Depa men o Ve e ina y Biosciences, Resea ch P og ams Uni , Molecula Neu ology, Uni e si y o Helsinki and Folkhälsan Resea ch Ins i u e, Helsinki, Finland, 5 Ag i ood Resea ch Finland, MTT, Bio echnology and Food Resea ch, Jokioinen, Finland, 6 Chai o Li es ock Bio echnology, Technische Uni e si ä München, F eising, Ge many Abs ac Impai ed mig a ion o p imo dial ge m cells du ing emb yonic de elopmen causes he edi a y gonadal hypoplasia in bo h sexes o No he n Finnca le and Swedish Moun ain ca le. The a ec ed gonads exhibi a lack o o , in a e cases, a educed numbe o ge m cells. Mos a ec ed animals p esen le -sided gonadal hypoplasia. Howe e , igh - sided and bila e al cases a e also ound. This ype o gonadal hypoplasia p e ails in animals wi h whi e coa colou . P e ious s udies indica ed ha gonadal hypoplasia is inhe i ed in an au osomal ecessi e ashion wi h incomple e pene ance. In o de o iden i y gene ic egions unde lying gonadal hypoplasia, a genome-wide associa ion s udy (GWAS) and a copy numbe a ia ion (CNV) analysis we e pe o med wi h 94 animals, including 21 a ec ed animals, using bo ine 777,962 SNP a ays. The GWAS and CNV esul s e ealed wo signi ican ly associa ed egions on bo ine ch omosomes (BTA) 29 and 6, espec i ely (P=2.19 x 10-13 and P=5.65 x 10-6). Subsequen cy ogene ic and PCR analyses demons a ed ha homozygosi y o a ~500 kb ch omosomal segmen ansloca ed om BTA6 o BTA29 (Cs29 allele) is he unde lying gene ic mechanism esponsible o gonadal hypoplasia. The duplica ed segmen includes he KIT gene ha is known o egula e he mig a ion o ge m cells and p ecu so s o melanocy es. This duplica ion is also one o he wo ansloca ions associa ed wi h colou sidedness in a ious ca le b eeds. Ci a ion: Venho an a H, Pausch H, Wysocki M, Szcze bal I, Hänninen R, e al. (2013) Ec opic KIT Copy Numbe Va ia ion Unde lies Impai ed Mig a ion o P imo dial Ge m Cells Associa ed wi h Gonadal Hypoplasia in Ca le (Bos au us). PLoS ONE 8(9): e75659. doi:10.1371/jou nal.pone.0075659 Edi o : Toshi Shioda, Massachuse s Gene al Hospi al, Uni ed S a es o Ame ica Recei ed Ma ch 8, 2013; Accep ed Augus 16, 2013; Published Sep embe 26, 2013 Copy igh : © 2013 Venho an a e al. 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, 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 au ho and sou ce a e c edi ed. Funding: Funding om he Academy o Finland, Finnish Ve e ina y Founda ion, O ion-Fa mos esea ch ounda ion, Emil Aal onen Founda ion and Niemi Founda ion is highly app ecia ed. HP was unded by he Ge man Fede al Minis y o Educa ion and Resea ch wi hin he Ag oClus E “Synb eed - Syne gis ic plan and animal b eeding”. The unde s had no ole in s udy design, da a collec ion and analysis, decision o publish, o p epa a ion o he manusc ip . Compe ing in e es s: The au ho s ha e decla ed ha no compe ing in e es s exis . * E-mail: [email p o ec ed] ☯ These au ho s con ibu ed equally o his wo k. In oduc ion Gonadal hypoplasia is cha ac e ised by abe an ly small and unde de eloped gonads. Fe ili y is gene ally dis u bed when bo h gonads a e hypoplas ic. Di e en ypes o gonadal hypoplasia ha e been epo ed in se e al mammalian species, including ca s, dogs [1], sheep [2], ho ses [3,4] and humans [5]. He edi a y gonadal hypoplasia is a equen diso de in wo Scandina ian ca le b eeds, No he n Finnca le and Swedish Moun ain ca le (also called Swedish Highland b eed). The de ec eme ged in he ea ly 20 h cen u y a e pu e-b eeding o Swedish Moun ain ca le was implemen ed [6,7]. A ew decades la e he incidence o gonadal hypoplasia inc eased hea ily and clinical in es iga ions we e ini ia ed. Comp ehensi e heal h con ol e alua ions we e implemen ed which educed he incidence o gonadal hypoplasia om 17.3% o Swedish Moun ain ca le bo n be o e 1937 o 7.3% o animals bo n be ween 1952 and 1954 [8]. The No he n Finnca le almos became ex inc du ing he Second Wo ld Wa . Thus conside able in og ession o genes om Swedish Moun ain ca le ook place du ing he las 65 yea s and likely in oduced gonadal hypoplasia in o he No he n Finnca le he d. PLOS ONE | www.plosone.o g 1 Sep embe 2013 | Volume 8 | Issue 9 | e75659 The incidence o gonadal hypoplasia in he men ioned b eeds is equal in bo h sexes and comp ehensi e b eeding expe imen s sugges a ecessi e mode o inhe i ance wi h incomple e pene ance [7]. Pene ance was es ima ed o be 0.5 and E iksson [7] sugges ed ha he incomple e pene ance is pa ly gene ically de e mined. Gonadal hypoplasia is unique in No he n Finnca le and Swedish Moun ain ca le because i is mainly mani es ed on he le side. The p opo ions o le -, double- and igh -sided gonadal hypoplasia a e 82%, 15% and 3%, espec i ely [7]. Al hough he absolu e numbe s di e ac oss s udies, le -sided gonadal hypoplasia always p edomina es [9,10]. Howe e , he e is no e idence ha he localiza ion (le and/o igh ) o hypoplas ic gonads is gene ically de e mined in No he n Finnca le and Swedish Moun ain ca le [7]. The se e i y o gonadal hypoplasia a ies conside ably om o al (Figu e 1) o pa ial [6,7] and a ec ed animals lack o ha e a educed numbe o ge m cells [6]. Bila e ally a ec ed animals a e o en s e ile and hypoplasia can also a ec seconda y sexual cha ac e is ics ia impai ed p oduc ion o sexual ho mones in emales [7,9]. The e a e no epo s ha a ec ed animals su e om any o he heal h p oblems. Gonadal hypoplasia o No he n Finnca le and Swedish Moun ain ca le is a congeni al de ec [7] and Se e g en [6] ound ha he hypoplas ic o a ies can al eady be iden i ied a he oe al s age. Foe al es icles we e no s udied. Se e g en Figu e 1. Le -sided gonadal hypoplasia. (A) Hypoplas ic o a y (le a ow) and no mal o a y ( igh a ow). (B) Hypoplas ic es icle (le ) and no mal es icle ( igh ). doi: 10.1371/jou nal.pone.0075659.g001 [6] concluded ha he eason o gonadal hypoplasia is a ailu e o he mig a ion and synch onous mi o ic di isions o p imo dial ge m cells (PGC). Mig a ion and p oli e a ion o PGCs in he de eloping emb yo is a complex p ocess wi h a ious genes and signalling pa hways in ol ed [11,12]. Gonadal hypoplasia occu s only in animals ha a e 60-100% whi e colou ed (Table S1) [6]. The coa colou o he No he n Finnca le and Swedish Moun ain ca le a ies om whi e o black o b own wi h nume ous in e media es. Whi e coa wi h pigmen ed ea s and muzzle is he mos common o m, bu spo ed sides and colou ed legs a e also commonly ound (Figu e S1). Pa o his colou a ia ion is likely due he colou - sided pa e n ha is sugges ed o be caused by a semi- dominan gene symbolized as Cs [13,14]. Recen ly Du kin e al. [15] showed ha colou sidedness is de e mined by wo loci p esen on bo ine ch omosomes, (BTA) 29 and 6. The Cs29 allele in BTA29 esul ed om a duplica ion and ansloca ion e en o a 492 kb segmen o BTA6 including he KIT gene. The Cs6 allele esiding on BTA6 is a esul o a subsequen duplica ion and ansloca ion e en ha mo ed back he segmen comp ising used sequences o he BTA29 and BTA6 o he KIT locus in BTA6. I is expec ed ha in bo h cases he dys egula ion o he KIT gene leads o he colou sidedness [15]. The KIT gene encodes a ype III ecep o p o ein o he y osine kinase amily. Se e al s udies ha e shown ha KIT p o ein is c ucial o su i al, p oli e a ion and mig a ion o melanocy e p ecu so s and p imo dial ge m cells (PGC) du ing emb yogenesis [16-18]. KIT is also essen ial o egula de elopmen o hema opoie ic cells [16,19]. In mice, mu a ions in KIT esul in impai ed pigmen a ion, educed e ili y o s e ili y, anaemia and dea ness. These mu a ions a e o en pleio opic (MGI:96677, h p://www.in o ma ics.jax.o g/). KIT mu a ions also unde lie coa colou a ia ion in o he mammals, e.g. pig [20,21] and ho se [22]. In humans, KIT mu a ions cause piebaldism, mas cell disease and se e al ypes o umou (*164920, h p://omim.o g/). The aim o his s udy was o iden i y he gene ic cause o he he edi a y o m o gonadal hypoplasia in No he n Finnca le and Swedish Moun ain ca le. A genome-wide associa ion s udy combined wi h a copy numbe a ia ion (CNV) analysis was pe o med wi h high-densi y SNP a ays o pinpoin he genomic egion. The esul s we e con i med wi h cy ogene ic in es iga ions, quan i a i e PCR and con en ional PCR. We show ha Swedish Moun ain ca le and No he n Finnca le ca y alleles ha cause colou -sidedness [15] and ha a ec ed indi iduals a e homozygous o he Cs29 allele ha includes he ec opic KIT gene. Resul s A genome-wide associa ion s udy maps congeni al gonadal hypoplasia o BTA29 Gonadal hypoplasia in No he n Finnca le and Swedish Moun ain ca le is hypo hesized o esul om pleio opic e ec s o a whi e colou gene because mos o he a ec ed animals a e p edominan ly whi e [6,23]. To accoun o bo h he di e en coa colou pa e ns and he a ec ion s a us o he He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 2 Sep embe 2013 | Volume 8 | Issue 9 | e75659 animals, ou di e en case-con ol coho s we e es ablished (Table 1) and allelic associa ions we e pe o med. P- alues below 7.71 x 10-8 (Bon e oni-co ec ed h eshold o mul iple es ing) we e conside ed o be signi ican ly associa ed. A genome-wide associa ion s udy (GWAS) wi h 20 una ec ed p edominan ly black and b own animals and 21 a ec ed p edominan ly whi e animals e ealed 18 signi ican ly associa ed SNPs on BTA29 be ween 17.69 Mb o 20.50 Mb (Figu e 2A, Table S2). The SNP showing mos signi ican associa ion was Bo ineHD2900005672 (19,661,149 bp, P=2.19 x 10-13). The second GWAS, including all a ailable una ec ed animals (N=73) and all a ec ed animals, yielded i e highly signi ican ly associa ed SNPs on BTA29 and one highly signi ican ly associa ed SNP on BTA23 a 52,435,290 bp, (P=9.57 x 10-9) (Figu e 2B). This SNP was conside ed o be a alse posi i e esul because he e we e no o he signi ican ly associa ed SNPs on BTA23. Mo eo e , e-geno yping he SNP using Sange -sequencing e ealed disc epan geno ypes compa ed wi h he geno ypes ob ained wi h he Illumina Bo ineHD Bead chip in some samples. This indica ed echnical p oblems. Compa ison o p edominan ly whi e -40 una ec ed and 21 a ec ed – animals e ealed a sugges i e associa ion (Table S2). Simila ly, GWAS in he ou h g oup (D) using 40 una ec ed p edominan ly whi e animals and 20 una ec ed p edominan ly colou ed animals e ealed a sugges i e associa ion (Table S2). Howe e he SNPs o he las wo GWAS analysis we e no signi ican ly associa ed a e Bon e oni-co ec ion (Figu e 2C and 2D). Addi ionally, associa ion analyses we e epea ed using a geno ypic es . The esul s we e in line wi h he allelic es , unde lining he impo ance o he 17.69 Mb o 20.50 Mb a ea in BTA29 (Table S3). Collec i ely, he analysis o he ou s udied coho s indica es ha he BTA29 egion is associa ed wi h gonadal hypoplasia and no only e lec s di e en coa colou pa e ns. The genomic in la ion ac o s o he g oups anged om 0.94 o 1.02 and indica e he absence o po en ial spu ious associa ions. CNV analysis iden i ies a ansloca ed genomic egion wi h he KIT gene Du kin e al. [15] showed ha colou sidedness in a ious ca le b eeds is de e mined by wo CNVs ha duplica e a small pa o he BTA6 and BTA29. Because o he s ong link be ween whi e colo a ion and gonadal hypoplasia in No he n Finnca le and Swedish Moun ain ca le, we ca ied ou CNV analysis. A o al o 2101 au osomal CNVs we e iden i ied in 94 animals. Among hem a CNV segmen on BTA6, ex ending om 71.60 Mb o 72.13 Mb and con aining he KIT gene, was signi ican ly associa ed wi h gonadal hypoplasia (P=5.65 x 10-6) (Figu e 3, Figu e 4A). The CNV was p esen in all 21 a ec ed animals and in 52 o 73 animals o he con ol g oup. Also, a CNV segmen on BTA29, ex ending om 19.86 Mb o 20.32 Mb, appea ed o be equen in he No he n Finnca le. The CNV segmen on BTA29 was iden i ied in 14 a ec ed and 44 una ec ed animals, bu was no associa ed wi h gonadal hypoplasia (Figu e S2). This segmen is in he immedia e icini y o he mos signi ican ly associa ed gonadal hypoplasia SNPs (Figu e 4B). To ensu e ha CNV segmen on BTA6 was no associa ed o gonadal hypoplasia nominal p- alues o he GWAS o he SNPs loca ed icini y o he segmen we e s udied (Tables S4 and S5). Only wi h he Design D in he geno ypic es a ew SNPs wi h sugges i e associa ion was ound. This design included only una ec ed p edominan ly whi e and colou ed animals so he ound associa ions we e ela ed o coa colou no gonadal hypoplasia. Du kin e al. [15] also showed ha he wo colou sidedness loci esul om a ansloca ion o a ch omosomal segmen , encompassing he KIT gene, om BTA6 o BTA29 and a pa o his agmen back o BTA6 nea he KIT gene. To in es iga e whe he simila duplica ions a e p esen in No he n Finnca le and he Swedish Moun ain b eed, we used 38 animals wi h copy numbe (CN) ou on BTA6 as he case g oup and 21 animals wi h CN wo as a con ol g oup o iden i y a po en ial ansloca ion. The CN s a us was de e mined based on he a ay geno ypes. We ound a s ong associa ion on BTA29 (19,661,149 bp, P = 1.66 x 10-28), indica ing a ansloca ion o he BTA6 segmen o BTA29 (Figu e S3). We nex analysed 58 and 36 animals wi h and wi hou CNV on BTA29, espec i ely, and iden i ied a s ong associa ion on BTA6 (72,198,048 bp, P=1.69 x 10-14, Figu e S4), indica ing a ansloca ion om he BTA29 segmen o BTA6. These posi ions ag ee wi h hose desc ibed by Du kin e al. [15] indica ing ha he duplica ions iden i ied a e he same. Table 1. O ganisa ion o ou di e en case-con ol designs. Con ol coho Case coho Gonadal hypoplasia UNAFF AFF UNAFF AFF Coa colou W B U W B U o al W B U W B U o al Design A 20 20 21 21 Design B 40 20 13 73 21 21 Design C 40 40 21 21 Design D 40 40 20 20 Fou di e en case-con ol coho s we e es ablished (A-D). The coa colou o he animals was assessed as p edominan ly whi e (W), p edominan ly black/b own (B) and unknown (U). The gonadal hypoplasia s a us was assessed as a bina y ai (a ec ed (AFF) and una ec ed (UNAFF)). doi: 10.1371/jou nal.pone.0075659. 001 He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 3 Sep embe 2013 | Volume 8 | Issue 9 | e75659 Figu e 2. Associa ion o 647,971 SNPs in ou di e en case con ol scena ios. The Manha an plo s ep esen he -log10(P) alues o associa ion o 647,971 SNPs in ou di e en case-con ol designs (Table 1). The ed do s ep esen signi ican ly associa ed SNPs (P < 7.71 x 10-8). doi: 10.1371/jou nal.pone.0075659.g002 Figu e 3. Associa ion o 2101 au osomal CNVs wi h he a ec ion s a us o 94 animals. The p esence o CNV segmen s was compa ed in 21 cases and 73 con ols using Fishe exac es s. The do s ep esen SNPs wi hin he CNV segmen s. Red do s ep esen SNPs in signi ican ly o e ep esen ed CNVs in cases s. con ols (P < 2.38 x 10-5). doi: 10.1371/jou nal.pone.0075659.g003 He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 4 Sep embe 2013 | Volume 8 | Issue 9 | e75659 Cy ogene ic s udies con i m ansloca ions be ween BTA6 and BTA29 Th ee No he n Finnca le animals we e selec ed o cy ogene ic in es iga ions based on he CN s a us o he BTA6- segmen . One o hese animals was a ec ed by gonadal hypoplasia. A Wes e n Finnca le and an Eas e n Finnca le animal we e also included in he compa a i e analysis. These b eeds a e no a ec ed by he gonadal hypoplasia. T adi ionally, Wes e n Finnca le is a solid b own b eed and Eas e n Finnca le is a b own b eed wi h he colou -sided pheno ype. Bac e ial a i icial ch omosome (BAC) clones speci ic o he CN a eas o ch omosomes BTA6 ( ed FISH signal) and BTA29 (g een FISH signal) we e used as p obes in FISH expe imen s. O e lapping ed and g een signals appea yellow. In he Wes e n Finnca le animal wo ed (BTA6 homologs) and wo g een (BTA29 homologs) signals we e eco ded. The Eas e n Finnca le animal showed he ollowing pa e n: wo yellow signals on BTA6 and wo g een signals on BTA29. All h ee animals om he No he n Finnca le b eed showed ed and yellow signals on BTA6 while he pa e ns on BTA29 a ied. The i s una ec ed No he n Finnca le showed g een and yellow signals on BTA29 whe eas he second una ec ed No he n Finnca le showed only g een signals. The a ec ed No he n Finnca le animal had yellow signals on BTA29 (Figu e S5). Taken oge he , all animals excep he solid b own Wes e n Finnca le animal had one o se e al ansloca ions associa ed wi h colou sidedness. The a ec ed No he n Finnca le was homozygous o he Cs29 allele. This suppo s ou GWAS and CNV esul s ha showed s ong associa ion be ween he Cs29 allele and gonadal hypoplasia. CNV con i ma ion using quan i a i e PCR The numbe o he Cs29 alleles was analysed by quan i a i e PCR in nine a ec ed and 21 una ec ed animals. The esul s ag ee wi h he CNV esul s based on he a ay geno ypes. Howe e , using qPCR we could no dis inguish be ween he e ozygous and homozygous animals. Figu e 4. Schema ic iew o wo CNV segmen s on BTA6 and BTA29. The igu es display 5-SNP-sliding-window log R a ios in 75 una ec ed (g ey) and 21 a ec ed (blue) animals on BTA6 (A) and BTA29 (B). The g ey shaded box ep esen s he ex en o he wo CNV segmen s. The gene con en was assessed based on he Uni e si y o Ma yland assembly. doi: 10.1371/jou nal.pone.0075659.g004 He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 5 Sep embe 2013 | Volume 8 | Issue 9 | e75659 Con i ma ion o CNV using con en ional PCR To cha ac e ize he ansloca ion p esen in No he n Finnca le and Swedish Moun ain ca le we used PCR p ime s acco ding o Du kin e al. [15] and one addi ional p ime pai ha lanked he inse ion si e o wild ype BTA29. Analysed animals included also wo animals ha we e omi ed om he GWAS analyses because o he geno yping ailu e. Colou o hese wo animals was p edominan ly b own and whi e. The esul s showed ha all a ec ed animals we e homozygous o he Cs29 allele and 15 animals had he Cs6 allele. O he una ec ed animals, 18 we e homozygous and 37 we e he e ozygous o he Cs29 allele and 20 did no ha e he Cs29 allele. The Cs6 allele was de ec ed om 44 con ol animals (Table 2). The a io o he Cs6 allele is sligh ly highe in a ec ed han in una ec ed animals: 15/21 (71%) and 44/75 (59%) espec i ely. Mo e impo an ly he occu ence o Cs6 in con ol animals ha we e homozygous o he Cs29 allele was 14/18 (78%) which is almos he same as in a ec ed animals (Table 2). Like he case animals, all homozygous con ol animals we e p edominan ly whi e. The a io o he Cs6 allele in whi e con ol animals ha we e he e ozygous o he Cs29 allele was 11/18 (61%) which is also qui e close o he Cs6 a io o he case animals (Table S6). These esul s sugges s ha he homozygous Cs29 allele migh ha e pleio opic e ec s on bo h coa colou and gonadal hypoplasia bu he Cs6 allele a ec s only he coa colou . The esul s a e in line wi h he CNV esul s based on a ay geno ypes and qPCR. Addi ionally, we de e mined he DNA sequence o all PCR p oduc s ampli ied om he selec ed 12 animals ( esul s no shown). The sequences we e compa able wi h hose ob ained by Du kin e al. [15]. A genome-wide associa ion s udy compa ing he case and con ol animals homozygous o he Cs29 allele e ealed no associa ion Gi en ha he Cs29 allele was associa ed wi h gonadal hypoplasia we pe o med an addi ional condi ional GWAS whe e he case animals we e compa ed wi h 18 con ol animals ha we e homozygous o he Cs29 allele. The only sugges i ely associa ed allele was he SNP on BTA23 a 52,435,290 bp which was shown o ha e disc epan geno ypes (Figu e S6). Table 2. The PCR analysis o CNVs. A ec ed Una ec ed +/+ +/Cs29 Cs29/Cs29 +/+ +/Cs29 Cs29/Cs29 +/+ 0 0 6 10 17 4 Cs6/- 0 0 15 10 20 14 o al 0 0 21 20 37 18 The PCR analysis o CNVs based on p ime s designed by Du kin e al. [15] and us. All 96 animals we e analysed including wo con ol animals whose geno yping ailed. The animals a e di ided acco ding o alleles in BTA29 and BTA6. He e ozygous and homozygous ca ie s o he Cs6 allele could no be dis inguished. doi: 10.1371/jou nal.pone.0075659. 002 Discussion The esul s o his s udy s ongly sugges ha he p edominan ly le -sided gonadal hypoplasia in No he n Finnca le and Swedish Moun ain ca le is associa ed wi h a complex ch omosomal ea angemen including a ~500kb duplica ed segmen om BTA6 which is ansloca ed o BTA29. This duplica ed segmen includes he en i e coding sequence o he KIT gene. Gi en ha he a ec ed animals we e homozygous o he ansloca ed duplica ion, ou s udy suppo s he ecessi e disease model and highligh s KIT gene dosage as a signi ican isk ac o . We showed ha he ansloca ed segmen is he same as one o he wo p e iously desc ibed ansloca ed segmen s Cs29 and Cs6 ha a e esponsible o colou sidedness in ca le [15]. The colou -sided pa e n is demons a ed by a iable coa ing om colou ed animals wi h a whi e line in he back o almos whi e animals wi h colou ed pa s only in ea s, muzzle and ee . The la e coa colou pa e n is common o No he n Finnca le and Swedish Moun ain ca le b eeds. Se e g en [6] showed ha he p edominan ly whi e coa colou o Swedish Moun ain ca le was s ongly associa ed wi h gonadal hypoplasia. Ou GWAS da a suppo he hypo hesis ha he Cs29 allele is associa ed wi h bo h whi e coa colou and gonadal hypoplasia. We did no ind any associa ion when compa ing p edominan ly whi e a ec ed and una ec ed animals. Mo eo e , he associa ion signal was s onge when a ec ed animals we e compa ed wi h colou ed animals han when a ec ed animals we e compa ed wi h all una ec ed animals. These esul s a e because mos o he whi e animals ca y a leas one copy o he Cs29 allele and his implica es co ela ion o Cs29 allele wi h bo h whi e coa colou and gonadal hypoplasia. Howe e , compa ison o una ec ed p edominan ly whi e animals and una ec ed p edominan ly colou ed animals showed no associa ion ha would ha e exceeded he genome-wide h eshold le el. This esul was su p ising bu can be explained wi h he knowing ha some o he una ec ed p edominan ly colou ed animals we e colou sided and pa o hem had he Cs29 allele. Mo eo e , some o he una ec ed p edominan ly whi e animals lacked he Cs29 allele. This indica es s onge co ela ion o Cs29 allele wi h gonadal hypoplasia han wi h p edominan ly whi e coa colou . Compa isons ac oss he s udy g oups s ongly sugges ha he Cs29 allele is associa ed wi h gonadal hypoplasia and no only e lec s di e en coa colou pa e ns. I is unlikely ha Cs6 allele is ela ed o gonadal hypoplasia gi en ha he CNV analyses showed no associa ion be ween he Cs6 allele and gonadal hypoplasia. Fu he mo e, he con en ional PCR esul s showed ha he Cs6 allele is only sligh ly en iched in a ec ed animals when compa ed wi h all una ec ed animals. Mo e impo an ly, he equency o he Cs6 allele was almos he same in case animals as in con ol animals ha we e homozygous o he Cs29 allele. Mu a ions wi hin he KIT gene cause whi e colou also in ho se and pig, bu he e a e no epo s o associa ed gonadal hypoplasia. On he o he hand, se e al mouse models ha bou ing KIT mu a ions show pleio opic e ec s causing educed e ili y and whi e colou ing. Many o hese mice He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 6 Sep embe 2013 | Volume 8 | Issue 9 | e75659 mu an s su e om anaemia, umou o ma ion and dea ness (MGI:96677, h p://www.in o ma ics.jax.o g/). We a e no awa e ha he a ec ed animals in ou s udy had any o he disease symp oms besides gonadal hypoplasia. Ye he same kind o gonadal hypoplasia, as in No he n Finnca le and Swedish Moun ain ca le, has no been obse ed in any mouse s udies. Only o he Nguni b eed (Bos indicus x Bos au us c oss) a e he e epo s o a he edi a y o m o gonadal hypoplasia esembling ha o m ound in No he n Finnca le and Swedish Moun ain ca le [24-26]. In e es ingly, he Nguni b eed also p esen s he colou -sided pheno ype. Du kin e al. [15] implica ed ha loci on BTA6 and BTA29 accoun o mos , i no all, colou sidedness in ca le. Mo eo e , B enig e al. [27] showed ha in Whi e Galloway ca le and Whi e Pa k ca le he p edominan ly whi e coa colou was a ibu ed o he Cs29 allele. This was also shown in ou s udy whe e only i e animals ou o 62 p edominan ly whi e animals did no ha e he Cs29 allele. Se e al ca le b eeds ha e his allele in BTA29, bu he edi a y gonadal hypoplasia has only been epo ed in No he n Finnca le, Swedish Moun ain ca le and Nguni ca le. This migh be because he p opo ion o animals homozygous o he Cs29 allele is oo small in se e al colou sided b eeds o his ype o ecessi e gonadal hypoplasia wi h incomple e pene ance o be diagnosed. Ano he hypo hesis o he gene ic cause o he gonadal hypoplasia is ha Swedish Moun ain ca le and No he n Finnca le ha e o he b eed speci ic changes in he genome in addi ion o he Cs29 allele. This conclusion implies ha Nguni migh ha e di e en mu a ions ha cause he gonadal hypoplasia. The GWAS s udies showed no o he eliable associa ion be ween he cases and con ols han he associa ion o BTA29. Howe e , he addi ional mu a ions migh emain unde ec ed because o he small sample size in he p esen s udy o because o he loca ion o he mu a ion. I mu a ion is wi hin he ansloca ed a ea o in co esponding a ea in he o iginal ch omosome he mapping wi h GWAS migh no wo k. Ou indings ha all a ec ed animals had wo ec opic KIT genes suppo he pos ula ed mode o ecessi e inhe i ance [7]. Ne e heless, 17 una ec ed animals also had wo ec opic KIT genes. This migh be explained by incomple e pene ance o by he p esence o an addi ional mu a ion o mu a ions besides he Cs29 allele. In conclusion, we showed ha a he edi a y o m o gonadal hypoplasia in Swedish Moun ain ca le and No he n Finnca le is associa ed wi h homozygosi y o he Cs29 allele. The e is need o u he in es iga ions o unde s and ully he mechanisms causing he diso de . Howe e , o ou knowledge, his is he i s epo indica ing a duplica ion o he KIT gene wi h p edominan ly le -sided gonadal hypoplasia in mammals. Ma e ials and Me hods E hics s a emen The blood sampling and clinical examina ions we e ca ied ou ollowing s anda d e e ina y p o ocols in Finland. The Animal E hics Commi ee o he S a e P o incial O ice o Sou he n Finland app o ed all animal wo k (ESAVI-2010-03428/Ym-23). Gonad examina ion and sampling Clinical examina ions we e done du ing a m isi s by expe ienced e e ina ians. Bulls and bull cal es we e palpa ed o es icle size and symme y. Female animals olde han 16 mon hs, excluding animals mo e han i e mon hs p egnan , we e s udied by o a ian palpa ion pe ec um. A e clinical examina ion, 9 ml o EDTA enous blood we e collec ed om he animals sui able o his s udy. Semen samples o a i icial insemina ion bulls included in he s udy we e also collec ed. F om he pos -mo em s udy animals’ gonads we e examined isually, palpa ed and weighed. His ological and issue samples o DNA ex ac ion we e aken om he collec ed gonads. His ological samples we e subjec ed o s anda d Bouin’s ixa ion and embedded in pa a in. Sec ions (5 µm) we e cu and s ained wi h haema oxylin-eosin (HE). Animals conside ed a ec ed by gonadal hypoplasia Cows and hei e s wi h one o a y o ex emely small size we e conside ed o be a ec ed by gonadal hypoplasia. Usually he hypoplas ic o a y was unde ec able by palpa ion pe ec um. Bulls and bull cal es we e conside ed o be a ec ed by unila e al gonadal hypoplasia i hei es icles clea ly di e ed in size, i.e. one es icle being mo e han wo imes la ge han he o he es icle in young cal es and o bulls’ one es icle being mo e han h ee imes la ge han he o he es icle. Bila e al gonadal hypoplasia was diagnosed only in one bull selec ed o semen collec ion. Bo h es icles we e e y small and he e we e no spe m cells in he ejacula e. The es icle his ology only showed Se oli cells and no spe ma ogonia in semini e ous ubules in he hypoplas ic es icles. When possible, he clinically diagnosed a ec ed animals we e e-examined du ing a second a m isi o he gonads we e s udied a e slaugh e . DNA ex ac ion Genomic DNA om blood samples was ex ac ed by au oma ic isola ion (Magne ic Sepa a ion Module I, Chemagen) and genomic DNA om issue and semen samples was ex ac ed using a comme cially a ailable Qiagen Ki (QIAamp DNA Mini Ki ). Ex ac ions o he blood and issue samples we e made acco ding o he manu ac u e ’s ins uc ions. The ex ac ion o DNA om semen samples was made acco ding o DNA Pu i ica ion om Tissues-p o ocol in QIAamp DNA Mini Ki handbook wi h some modi ica ions. F om 200 o 500 µl o ozen semen was cen i uged o 5 min a 100 × g. The supe na an was mo ed and he pelle was washed wi h 200 µl o phospha e bu e ed saline (PBS). The Qiagen bu e ALT was added up o 300 µl and 20 µl o P o einase K and di hio h ei ol was also added. The mix was incuba ed o 1 h a 56 °C. Du ing incuba ion he sample was pulse o exed ou imes o 15 sec. 300 µl o Qiagen bu e AL was added and he pulse o ex epea ed. The sample was incuba ed o 10 min a 56 °C and he ea e 150 µl o 96% alcohol was added. The sample was pulse o exed and incuba ed o 3 min a oom empe a u e. The whole mix u e was applied o he QIAamp Mini spin column and cen i uged a 6,000 × g o 1 min. The He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 7 Sep embe 2013 | Volume 8 | Issue 9 | e75659 il a e was disca ded and he column was washed wice wi h 500 µl o Qiagen bu e AW1 and once wi h 500 µl o Qiagen bu e AW2 (cen i uga ion a 6,000 × g o 1 min). The column was d ied wi h cen i uga ion a 20,000 × g o 3 min. The DNA was elu ed wi h 50 µl o dis illed wa e , incuba ed o 1 min a oom empe a u e and cen i uged a 20,000 × g o 1 min. Animals selec ed o geno yping DNA samples om 96 animals we e included in his s udy. Animals wi h o al unila e al o bila e al hypoplasia we e included in he case g oup ha comp ised 21 animals (10 males and 11 emales). O hese animals, one was a ec ed wi h bila e al, 18 wi h le -sided and wo wi h igh -sided gonadal hypoplasia. The con ol g oup included 75 una ec ed animals. Among he s udied animals 91 we e No he n Finnca le and i e we e Swedish Moun ain ca le ( wo cases and h ee con ols). High-densi y geno yping and quali y con ol Nine y-six animals (21 a ec ed / 75 una ec ed) we e geno yped wi h he Illumina Bo ineHD Bead chip in e oga ing geno ypes o 777,962 SNPs. Geno ype calling was pe o med using de aul pa ame e s o Illumina’s BeadS udio. Quali y con ol was ca ied ou wi h PLINK 1.07 [28]. We excluded 1224, 343 and 1735 SNPs wi h Y-ch omosomal, mi ochond ial and unknown ch omosomal posi ion, espec i ely, o u he analysis. The geno ypes o wo una ec ed animals we e omi ed because geno yping ailed in mo e han 10% o he SNPs. We u he excluded 6229 SNPs because geno yping ailed in mo e 10% o he indi iduals and 121,657 monomo phic SNPs. The inal da ase comp ised 94 animals and 647,971 SNPs wi h an a e age call- a e o 99.67%. The ch omosomal posi ion o he SNPs was de e mined based on he UMD3.1 assembly o he bo ine genome [29]. Genome-wide associa ion s udy Fishe exac es s o allelic and geno ypic associa ions we e pe o med o compa e geno ypes in cases s. con ols a each SNP in u n using PLINK [28]. We conside ed SNPs wi h P < 7.71 x 10-8 as signi ican ly associa ed (Bon e oni-co ec ed h eshold o mul iple es ing). Quan ile-quan ile plo s we e inspec ed and genomic in la ion ac o s we e calcula ed acco ding o De lin and Roede [30] o assess he ex en o alse posi i e associa ion signals. De ec ion o copy numbe a ian s Geno ype signal in ensi ies ob ained om geno yping wi h he Illumina Bo ineHD Bead chip (see abo e) we e analysed wi h PennCNV [31] o iden i y copy numbe a ia ions (CNV). B ie ly, he implemen ed CNV-de ec ion algo i hm conside s bo h he log R a io (LRR) and he B allele equency (BAF), as well as he allele equency and he dis ance o adjacen SNPs. Two indi iduals and 6229 SNPs wi h poo geno yping quali y (see abo e) we e no conside ed o he iden i ica ion o CNVs. The p esence o CNVs con aining a minimum numbe o 10 SNPs co esponding o a minimum leng h o app oxima ely 35 Kb was compa ed in cases s. con ols using Fishe exac es s. Cy ogene ic analysis Me aphase sp eads we e ob ained om sho - e m lymphocy e cul u es, es ablished o i e animals. The FISH s udy was ca ied ou acco ding o a p o ocol desc ibed by Du kin e al. [15]. Two BAC clones (RP42-160M9, RP42-156I13) co e ing he egion o in e es on ch omosome BTA6 ( egion 72,566,605–72,817,995bp, UMD 3.0) and wo BAC clones (RP42-37P11, RP42-116G8) co e ing he egion o in e es on BTA29 ( egion 20,772,406–21,035,251bp, UMD 3.0) we e de i ed om he RPCI-42 Bo ine BAC Lib a y (h p:// bacpac.cho i.o g/home.h m). BAC DNA was isola ed using an alkaline lysis me hod and labelled by andom p iming. The isola ed DNAs om wo BAC clones speci ic o BTA6 we e mixed equally and labelled using bio in-11-dUTP, while DNA om wo BAC clones speci ic o BTA29 was also mixed and labelled wi h digoxigenin-11-dUTP. The labelled p obes wi h an excess o bo ine Co -1 DNA we e sepa a ely dena u ed o 10 min a 70°C and applied on dena u ed ch omosome slides. Hyb idiza ion was ca ied ou o e nigh a 37°C. A e slide washing, bio in-labelled p obes we e de ec ed using s ep a idin-Cy3 (Ame sham, 1:200, ed colou ) and digoxigenin-labelled p obes we e de ec ed wi h an idigoxigenin- luo escein Fab agmen s (Roche, 1:200, g een colou ). The slides we e coun e s ained wi h Vec ashield con aining DAPI (Vec o Labo a o ies) and examined wi h an epi luo escence Nikon E600 Eclipse mic oscope equipped wi h a cooled digi al CCD came a and Lucia so wa e. QPCR QPCR was used o alida e CNV iden i ied a e bioin o ma ics analysis. Al oge he 30 samples we e analysed: nine cases and 21 con ols. A simple me hod based on Weksbe e al. [32] and Lachman e al. [33] was applied o analysis. Two p ime pai s designed using P ime 3 [34] we e loca ed in he KIT egion: p ecisely in exon 8 (GGGCCAGTGGATGTACAGAT) and 9 (TGCAAAGTTAAAAGAGGCAGA); he second pai in exon 18 (CACATTTGAAAGTGATGTCTGG) and 19 (AGAACTTAGAATCGACTGGCATT). 10 µl o QPCR eac ion was composed om he Fas SYBR® G een Mas e Mix (Li e Technologies) and 2.5 pmol o each p ime . All ampli ica ions we e ca ied ou in Applied Biosys ems 7500 Fas Real-Time PCR Sys em (Li e Technologies) acco ding o manu ac u e ’s ecommenda ions. H6PD was used as a e e ence gene; bo h p ime s we e loca ed in he i s exon (AAGGTCCTGGAGTCCCTGTC, GTAGAAAATTCGGCCGGTCT). PCR and sequencing All animals we e analysed wi h s anda d PCR ca ied ou using b eakpoin p ime s designed by Du kin e al. (Table 2) [15] and one addi ional p ime pai PSK_α-β2_R (TGGGTAGACAGGTTTGTTTCC and TCTTGACCACTTGCATTGGA) ha lanked he inse ion si e o he wild ype BTA29. A PCR eac ion o 20 µl olume He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 8 Sep embe 2013 | Volume 8 | Issue 9 | e75659 con aining 20 ng genomic DNA, 1x Qiagen PCR bu e , 1.5 mM MgCl2, 200 µM o each nucleo ide, 5 pmol o each o wa d and e e se p ime (Sigma) and 0.5 uni s o Taq Polyme ase (Qiagen) was pe o med unde he ollowing condi ions: ini ial dena u ing a 95 °C o 3 min, ollowed by 35 cycles a 94 °C o 30 sec, 60 °C o 1 min, 72 °C o one minu e and he inal ex ension a 72 °C o 3 min. PCR p oduc s we e isualized oge he wi h GeneRule TM100bp DNA Ladde (Fe men as, The mo Scien i ic) on 1.5% aga ose gel. Fi e case samples, wo con ol samples wi h a duplica ion o he BTA6 segmen , wo con ol samples wi h bo h s udied duplica ions and h ee con ol samples wi hou he duplica ions we e subjec ed o sequencing. A 10-20 ng o pu i ied PCR p oduc was mixed wi h 0.5 µl BigDye® Te mina o 3.1 (Li e Technologies) and 0.5 µl o he o wa d o e e se PCR p ime (2.5 pmol). The sequencing eac ion was pe o med unde he ollowing condi ions: ini ial dena u ing a 96 °C o 10 sec, ollowed by 35 cycles, including 10 sec in 96 °C, 5 sec in 50 °C, and 4 min in 60 °C. The gel il a ion o he sequencing eac ion was applied using he Mul iSc een il a ion pla e (MAHVN4510; Millipo e) and Sephadex G-50 Fine (Sigma) ollowed by capilla y elec opho esis ca ied ou on 3130xl Gene ic Analyze (Li e Technologies). Base calling, sequence alignmen and polymo phism de ec ion we e made using he Ph ed/Ph ap/Polyph ed so wa e [35-37]. Sequences we e inspec ed using Consed [38]. Se en cases and en con ols we e sequenced (see abo e) wi h p ime s CAATTTTAAGCATGTGCTGAGG and ACAGCCTCTGGTCTGTCTGG o alida e he geno ype calls o he SNP on BTA23 a 52,435,290 bp. Suppo ing In o ma ion Figu e S1. Examples o he common colou pa e n in No he n Finnca le. Mos commonly, No he n Finnca le is almos whi e wi h black o b own in ea s and muzzle. The lanks and legs can also be pa ly colou ed o spo ed. (TIF) Figu e S2. A e age log R a io o animals ca ying he ec opic BTA29 segmen . The a e age log R a io was calcula ed om 15 a ec ed and 44 una ec ed animals ha ca y he duplica ed segmen o BTA29. The 5-SNP-sliding window log R a io is p esen ed o 563 SNPs. (PNG) Figu e S3. Localisa ion o a ansloca ion o a KIT con aining segmen o ch omosome 29. Animals ca ying wo and ou copies o he BTA6 segmen we e compa ed using Fishe exac es s. The ed do s ep esen signi ican ly associa ed SNPs (P < 7.71 x 10-8). (EPS) Figu e S4. Localisa ion o a ansloca ion o a BTA29 segmen o ch omosome 6. Animals wi h and wi hou he p esence o a BTA29 CNV we e compa ed using Fishe exac es s. The ed do s ep esen signi ican ly associa ed SNPs (P < 7.71 x 10-8). (EPS) Figu e S5. FISH s udies. Th ee No he n Finnca le (NFC) animals wi h di e en combina ions o he Cs29 allele and one animal o he Wes e n Finnca le (WFC) and Eas e n Finnca le (EFC) we e analysed by FISH wi h wo BAC p obes. The Cs29 allele is associa ed wi h bo h colou sidedness and gonadal hypoplasia and i co esponds o he ed FISH signal o he ed ba . The Cs6 allele is associa ed wi h colou sidedness and co esponds o he g een FISH signal o he g een ba . O e lapping ed and g een signals appea yellow. All animals excep he solid b own Wes e n Finnca le had one o se e al Cs alleles. The animal NFC 164 is a ec ed wi h gonadal hypoplasia and i is homozygous o he Cs29 allele. (TIF) Figu e S6. Associa ion o 647,971 SNPs wi h he a ec ion s a us o 39 animals homozygous o he Cs29 allele. Associa ion analysis was pe o med using Fishe exac es s o allelic associa ion o 21 a ec ed and 18 una ec ed animals homozygous o he Cs29 allele. (JPG) Table S1. The associa ion be ween he p opo ion o coa pigmen a ion and o al (unila e al o bila e al) gonadal hypoplasia in he Swedish Moun ain b eed emales (modi ied om Se e g en [6]). (DOCX) Table S2. Nominal p- alues o he SNPs signi ican ly associa ed o gonadal hypoplasia in BTA29 (allelic es ). P- alues we e calcula ed wi h Fishe exac es s in PLINK o de e mine allelic associa ion in ou di e en case-con ol coho s (Table 1). (XLSX) Table S3. Nominal p- alues o he SNPs signi ican ly associa ed o gonadal hypoplasia in BTA29 (geno ypic es ). P- alues we e ob ained using a 2d geno ypic es implemen ed in PLINK o de e mine associa ion in ou di e en case-con ol coho s (Table 1). (XLSX) Table S4. Nominal p- alues o he SNPs loca ed be ween 71502659 bp and 71990541 bp in BTA6 (allelic es ). P- alues we e calcula ed wi h Fishe exac es s in PLINK o de e mine allelic associa ion in ou di e en case-con ol coho s (Table 1). (XLS) Table S5. Nominal p- alues o he SNPs loca ed be ween 71502659 bp and 71990541 bp in BTA6 (geno ypic es ). P- alues we e ob ained using a 2d geno ypic es implemen ed in PLINK o de e mine associa ion in ou di e en case-con ol coho s (Table 1). (XLS) He edi a y Gonadal Hypoplasia in Ca le PLOS ONE | www.plosone.o g 9 Sep embe 2013 | Volume 8 | Issue 9 | e75659