Divergent evolution in the cytoplasmic domains of PRLR and GHR genes in Artiodactyla

BioMed Cen al Page 1 o 11 (page numbe no o ci a ion pu poses) BMC E olu iona y Biology Open Access Resea ch a icle Di e gen e olu ion in he cy oplasmic domains o PRLR and GHR genes in A iodac yla Te hi Iso-Tou u*1, Juha Kan anen1, Meng-Hua Li1,3, Zygmun Gizejewski2 and Johanna Vilkki1 Add ess: 1Bio echnology and Food Resea ch, MTT Ag i ood Resea ch Finland, 31600 Jokioinen, Finland, 2Ins i u e o Animal Rep oduc ion and Food Resea ch, Polish Academy o Sciences, Pl-10-747 Olsz yn, Tuwima 10, Poland and 3Ecological Gene ics Resea ch Uni , Depa men o Biological and En i onmen al Sciences, PO Box 65, FI-00014 Uni e si y o Helsinki, Finland Email: Te hi Iso-Tou u* - [email p o ec ed]; Juha Kan anen - j[email p o ec ed]; Meng-Hua Li - menghua.[email p o ec ed]; Zygmun Gizejewski - [email p o ec ed]; Johanna Vilkki - [email p o ec ed] * Co esponding au ho Abs ac Backg ound: P olac in ecep o (PRLR) and g ow h ho mone ecep o (GHR) belong o he la ge supe amily o class 1 cy okine ecep o s. Bo h o hem ha e been iden i ied as candida e genes a ec ing key quan i a i e ai s, like g ow h and ep oduc ion in li es ock. We ha e p e iously s udied he molecula ana omy o he cy oplasmic domain o GHR in di e en ca le b eeds and a iodac yl species. In his s udy we ha e analysed he co esponding cy oplasmic signalling egion o PRLR. Resul s: We sequenced PRLR gene exon 10, coding o he majo pa o he cy oplasmic domain, om ca le, Ame ican bison, Eu opean bison, yak, sheep, pig and wild boa indi iduals. We ound di e en pa e ns o a ia ion in he wo ecep o s wi hin and be ween uminan s and pigs. Pigs and bison species ha e no a ia ion wi hin GHR exon 10, bu show high haplo ype di e si y o he PRLR exon 10. In ca le, PRLR shows lowe di e si y han GHR. The Bo inae PRLR haplo ype ne wo k i s be e he known phylogene ic ela ionships be ween he species han ha o he GHR, whe e di e ences wi hin ca le b eeds a e la ge han be ween he di e en species in he sub amily. By compa ison wi h he wild boa haplo ypes, a high numbe o subsequen nonsynonymous subs i u ions seem o ha e accumula ed in he pig PRLR exon 10 a e domes ica ion. Conclusion: Bo h genes a ec a mul i ude o ai s ha ha e been a ge s o selec ion a e domes ica ion. The genes seem o ha e esponded di e en ly o di e en selec ion p essu es imposed by human a i icial selec ion. The esul s sugges possible e ec s o selec i e sweeps in GHR be o e domes ica ion in he pig lineage o species di e gence in he Bison lineage. The PRLR esul s may be explained by s ong di ec ional selec ion in pigs o unc ional swi ching. Backg ound Domes ica ion o he unique o m o mu ualism ha de elops be ween a human popula ion and a a ge ani- mal popula ion has s ong selec i e ad an ages o bo h pa ne s [1]. Li es ock domes ica ion s a ed a ound 10000 yea s ago in he Fe ile C escen , beginning wi h Published: 22 July 2009 BMC E olu iona y Biology 2009, 9:172 doi:10.1186/1471-2148-9-172 Recei ed: 11 May 2009 Accep ed: 22 July 2009 This a icle is a ailable om: h p://www.biomedcen al.com/1471-2148/9/172 © 2009 Iso-Tou u 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. BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 2 o 11 (page numbe no o ci a ion pu poses) goa s and sheep [2]. Impo an ai s ega ding domes ica- ion include beha iou and ep oduc ion as well as dai y- ing [3]. Milking o he uminan animals was p ac iced in ensi ely al eady o e 8000 yea s ago (six h and se en h millennia BC) in no hwes Ana olia. Domes ica ed li e- s ock species ha e unde gone and a e cons an ly unde a i icial selec ion. S ong di ec ional selec ion in domes- ic animals is pos ula ed o ha e led o selec i e sweeps in which alleles a loci ha unde lie selec ed ai s (g ow h, e ili y, milk p oduc ion, coa colou ) ha e dec eased o inc eased ma kedly in hei equency [4]. A ecen genome wide analysis o ca le iden i ied de ec able signa- u es o domes ica ion and a i icial selec ion in he ca le genome, bu also ha he cu en le els o di e si y wi hin b eeds a e a leas as g ea as wi hin humans [5]. Howe e , he e ec s o domes ica ion and a i icial selec ion a e s ill mos ly unknown a he le el o nucleo ide sequence a i- a ion. Accumula ing knowledge o he domes ic species genomes has enabled mapping o loci a ec ing key ai s unde a i icial selec ion in li es ock. In ca le, nume ous such quan i a i e ai loci (QTL) ha e been iden i ied. O e i y QTLs ha e been localized o ca le ch omosome 20 h p://genomes.sapac.edu.au/bo ineq l/. Two in e es - ing candida e genes wi h po en ial e ec s on a ious ag o- nomically impo an ai s, p olac in ecep o (PRLR) and g ow h ho mone ecep o (GHR) loca e in his ch omo- some a a dis ance o 7.5 Mb (B au4.0, Oc .2007, h p:// www.ensembl.o g). An S o N subs i u ion (S18N) in he signal pep ide o he PRLR is linked wi h p o ein and a yield [6] and he subs i u ion F279N in he ansmem- b ane pa o he GHR has been sugges ed o be a quan i- a i e ai nucleo ide/causa i e mu a ion a ec ing milk a and p o ein pe cen age [7]. Bo h GHR and PRLR belong o he supe amily o class I cy okine ecep o s, which p esumably a ose as he esul o mul iple gene duplica ions and subsequen di e gen e olu ion. GHR and PRLR sha e a common e ia y s uc- u e (an ex acellula domain, a single memb ane span- ning ansmemb ane domain, and a cy oplasmic domain). They media e he signals o hei ligands, he g ow h ho mone (GH), p olac in (PRL) and placen al lac- ogen (PL). The binding o he ligand o he ex acellula pa induces homo- o he e odime iza ion o he ecep- o s, ollowed by in acellula signal ansmission by he JAK-S a signalling pa hway [8,9]. Bo h GHR and PRLR a e in ol ed in mamma y g ow h and unc ion. The main biological ole o he GH is he con ol o pos na al g ow h, whe eas he addi ional epo ed e ec s o PRL include in ol emen in seasonali y, ep oduc ion, beha - iou and immuno egula ion. Amino acid sequence iden- i y be ween GHR and PRLR a ies be ween 30% and 70% depending on he pa o he ecep o . The g ea es simi- la i y o he aa sequences be ween GHR and PRLR is in he ex acellula domains. The cy oplasmic domain and espe- cially he well conse ed BOX1 is essen ial o signal ans- duc ion. We ha e p e iously examined he GHR cy oplasmic domain sequence in di e en ca le b eeds and A iodac- yla species [10], whe e we obse ed in e es ing polymo - phism. The aim o his s udy was o cha ac e ize pa e ns o polymo phism in he co esponding in acellula egion o PRLR in di e en A iodac yla species and b eeds o enable he compa ison o sequence e olu ion in esponse o domes ica ion and a i icial selec ion in wo e olu iona y ela ed and closely loca ed candida e genes. Resul s The sequence analysis o he cy oplasmic domain o he PRLR gene e ealed in e es ing a ia ion in wild and domes ica ed A iodac yla species and subspecies (Table 1). Al oge he 13 SNPs we e ound om he ca le sam- ples. O hese six we e p esen in bo h Eu opean and A i- can ca le: one nsSNP (E378K) and i e sSNPs (N 1088, N 1427, N 1622, N 1754, and N 1775). Eu opean ca le had wo p i a e nsSNPs (P340T, A536V) and wo p i a e sSNPs (N 1682, N 1817) while A ican ca le had wo p i- a e nsSNPs (V439M and L497R) and one p i a e sSNP (N 1769). The ca le SNPs we e in e ed o seg ega e as 14 haplo ypes (Figu e 1) wo o which we e sha ed by Eu o- pean and A ican ca le (haplo ypes BOS_PRLR1 and BOS_PRLR2). Ame ican bison and Eu opean bison sha ed one nsSNP (E384K). Besides ha , Ame ican bison had one p i a e nsSNP (D588E). Eu opean bison samples we e mo e di e gen ; hey had wo p i a e nsSNPs (Q397K and M446V) and one sSNP (N 1730). S a is i- cally in e ed haplo ypes o he Ame ican bison and Eu opean bison a e gi en in Figu e 1. Ame ican bison and Eu opean bison sha e one haplo ype, BBI_PRLR3/ BBO_PRLR3. The s udied yak samples we e monomo - phic (Figu e 1). Sheep samples ca ied ou sSNP (N 1007, N 1160, N 1217, and N 1400) and h ee nsSNP (E387K, A476T, and S480R) in exon 10. The E387K posi- ion has been ixed o E in he Bos lineage and K in he Bison lineage. S a is ical haplo ype econs uc ion unco e ed 4 di e en sheep haplo ypes. Cloning o one ambiguous sheep indi- idual e ealed wo unique haplo ypes (OVIS_PRLR5 and OVIS_PRLR6) and hese haplo ypes we e included in u - he analysis (Figu e 2). Numbe ing o he Bos and Bison, and O is SNPs is based on he e e ence sequence NM_001039726. All Bos, Bison and O is SNPs we e in Ha dy-Weinbe g equilib ium. F om he domes ic pig samples we ound one sSNP (N 1620) and eigh nsSNPs (L406P, D428A, A461G, BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 3 o 11 (page numbe no o ci a ion pu poses) K480R, M510L, G534S, G597S, and A601V). Numbe ing o he pig SNPs is based on he pig cDNA sequence DQ157757, acco ding o he numbe ing o p e iously iden i ied SNPs [11]. All SNPs we e in Ha dy-Weinbe g equilib ium. The SNPs in pig exon 10 cons i u e 5 di e - en haplo ypes (Figu e 3). Se en ou o nine wild boa indi iduals we e SUS_PRLR4 homozygo es, one indi id- ual was SUS_PRLR5 homozygo e and one indi idual had haplo ype SUS_PRLR4 and haplo ype SUS_PRLR4-del. SUS_PRLR4-del haplo ype has a h ee-nucleo ide dele ion (N 1439 – N 1441) inducing lack o aa 480. PRLR haplo- ype sequences om he di e en species ha e been deposi ed in GenBank h p://www.ncbi.nlm.nih.go wi h he accession numbe s FJ901275 – FJ901301 and FJ901303 – FJ901307. Compa ison o a ia ion in he GHR and PRLR cy oplas- mic domains in he species analysed in bo h s udies is p e- sen ed in Table 1. New esul s om he wild boa and Eu opean bison GHR sequencing ha e been added. All wild boa s we e monomo phic ha ing he same haplo- ype in GHR exon 10 as domes ic pigs (FJ901302) and Eu opean bison was shown o ha e he same GHR exon 10 haplo ype [GenBank:DQ062723] as Ame ican bison. The mos d ama ic di e ence in he a ia ion be ween GHR and PRLR wi hin species is seen in he domes ic pig. The comple e lack o a ia ion in GHR is con as ed wi h he high haplo ype and nucleo ide di e si y in he PRLR, consis ing o mainly nonsynonymous a ia ion. In he Bo inae sub amily PRLR and GHR show di e en ea u es in he wo Bison species om ca le. The Bison spe- cies, like pigs, a e monomo phic o GHR; bu show high le el o haplo ype di e si y in PRLR. The A ican ca le shows equally high le els o di e si y a bo h genes, whe eas in Eu opean ca le he haplo ype and nucleo ide di e si y is lowe a PRLR. The yak is monomo phic o bo h genes. Sheep belongs o he same amily, Bo idae, as Bo inae species. Sheep has simila numbe o nsSNPs in GHR as in PRLR; howe e PRLR in all is mo e polymo - phic. The di e en indices o neu ali y a PRLR, Tajima's D alue, Fu and Li's D* and F* alues a e mainly e lec ing he same pa e n wi h each o he . The D alues o A ican ca le, Eu opean bison and domes ic pig a e posi i e and o Eu opean ca le and sheep nega i e (Table 1). Only he alues o A ican ca le (bo h a PRLR and GHR) de ia e s a is ically signi ican ly om ze o. A i icially selec ed popula ions, like li es ock species, do no ul il he assump ions o andom ma ing and cons an popula ion size o he neu ali y es , hence posi i e Tajima's D al- ues a e likely due o he demog aphic his o ies o hese species o b eeds a he han ue balancing selec ion. Conside ing he known his o y o zebu- au ine c oss- b eeding in A ica [12], he obse ed allelic di e si y is mos likely caused by admix u e A sliding window plo o he nucleo ide di e gence along PRLR exon 10 in 18 e eb a e species is p esen ed in Fig- u e 4. Two egions o low nucleo ide di e si y a e e iden , one a he beginning and one in he egion be ween 270 bp and 400 bp om he beginning o he exon 10. All E o K amino acid changes in Bo idae (ca le E378K, Ame i- Table 1: Di e si y and neu ali y indices in exon 10 o PRLR and GHR genes in he s udied species. Gene Species No. o No. o Tajima's Fu and Li's n lengh (bp) ns s haplo ypes HdS±T±DD*F* PRLR Eu opean Ca le (Bos au us) 216 891 3 7 9 0.44 0.84 1.69 -1.08 -0.41 -0.8 A ican Ca le (Bos indicus) 22 891 3 6 7 0.76 3.98 2.32 2.05* 1.36 1.86* Yak (Bos g unniens) 4891 1 Ame ican bison (Bison bison) 3 891 2 0 3 0.73 1.05 0.98 0.31 0.06 0.12 Eu opean bison (Bison bonasus) 5 891 3 1 3 0.73 2.17 1.59 1.41 1.24 1.43 Sheep (O is a ies) 14 891 3 4 6 0.61 1.18 1.4 -0.47 -0.78 -0.8 Domes ic pig (Sus sc o a) 18 750 8 1 5 0.63 3.27 2.89 0.39 1.36 1.24 Wild boa (Sus Sc o a)9750112 0.21 0.56 0.78 -0.68 0.88 0.53 GHR Eu opean Ca le (Bos au us) 202 900 4 3 18 0.69 2.07 1.18 1.47 1.09 1.47 A ican Ca le (Bos indicus) 47 900 4 6 7 0.72 3.97 2.17 2.13* 1.38 1.94* Yak (Bos g unniens) 4 900 1 Ame ican bison (Bison bison)3900 1 Eu opean bison (Bison bonasus) 6900 1 Sheep (O is a ies) 14 900 3 4 0.66 1.18 0.86 0.92 0.96 1.1 Domes ic pig (Sus sc o a)18690 1 Wild boa (Sus Sc o a)9690 1 Eu opean ca le b eeds a e g ouped and A ican ca le b eeds a e g ouped. ns = nsSNP, s = sSNP, Hd = haplo ype di e si y, S = nucleo ide di e si y, TW = Wa e son's he a es ima o , * P < 0.05, ± pe kilobase be ween sequences, does no include SUS_PRLRL4_del haplo ype BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 4 o 11 (page numbe no o ci a ion pu poses) can/Eu opean bison E384K, and sheep E387K) a e wi hin 27 bp in he second low di e si y module and addi ion- ally one Eu opean bison K o Q subs i u ion and wo pig subs i u ions (L406P and D428A) all inside his egion. We s udied he possible impac s o he aa changes o he p o ein s uc u e wi h he me hods implemen ed in SIFT and PolyPhen p og ams [13,14]. The subs i u ion E378K ound in ca le was p edic ed o a ec p o ein unc ion by SIFT analysis (sco e 0.02). Two o he SNPs, E384K in bo h bison species and V439M in ca le, go SIFT sco es below 0.1 (Table 2). This cu -o alue has been sugges ed o p o- ide be e sensi i i y o de ec ing dele e ious SNPs [13]. E (glu amic acid) is a pola , acidic amino acid; K (lysine) is a pola and basic amino acid; V ( aline) and M (me hio- nine) a e nonpola and neu al amino acids. Linkage disequilib ium (LD) analysis e ealed ha in Eu opean ca le only one pai wise SNP compa ison showed LD ( 2 = 1). On he o he hand, in A ican ca le se en pai wise SNP compa isons we e in LD ( 2 = 1). The same LD pa e n was isible in GHR exon 10 [10]. In sheep one pai wise SNP compa ison was in LD ( 2 = 1), in pigs wo pai wise SNP compa isons showed LD ( 2 = 1). A pa simony haplo ype ne wo k o he sub amily Bo i- nae is p esen ed in Figu e 5a. Mos o he common ca le haplo ypes di e om he majo haplo ype BOS_PRLR1 only by one nucleo ide. Howe e , he e a e se e al deduced in e media e haplo ypes ha we e no seen in his s udy (and some o he subs i u ions a e p esen in wo di e en loca ions on he ne wo k). The mos com- mon haplo ype o he A ican ca le is BOS_PRLR4. I is absen in Eu opean ca le and di e s by wo nonsynony- mous and i e synonymous subs i u ions om he BOS_PRLR1 ( he mos common haplo ype o he Eu o- pean ca le). Thus he BOS_PRLR4 as well as BOS_PRLR9, BOS_PRLR10, BOS_PRLR11 and BOS_PRLR12 p obably ep esen zebu haplo ypes in ou da a, e lec ing he deep di e gence be ween he au ine and zebu genomes [12]. Acco ding o he PRLR ne wo k, he yaks a e mo e closely ela ed o he genus Bison han o he genus Bos. The mos equen ca le haplo ype BOS_PRLR1 and he yak haplo- S a is ically in e ed ca le, yak, Ame ican bison and Eu opean bison haplo ypes om he PRLR exon 10Figu e 1 S a is ically in e ed ca le, yak, Ame ican bison and Eu opean bison haplo ypes om he PRLR exon 10. Only a iable si es a e shown. SNP numbe ing is based on he sequence NM_001039726.1. Amino acid numbe ing s a s om he i s me hionine in he p o ein sequence NP_001034815.1.                                 !!!"#$ %  &           '$())))*$ + $ ! ,-./+%+ 00000 0000000 ,-./+%+00000 0000000 ,-./+%+0000 0000000 ,-./+%+000000000 ,-./+%+00000 000000 ,-./+%+ 00000 00000000 ,-./+%+ 0000 0000000 ,-./+%+ 00000 0000000 ,-./+%+ 0000 0000000 ,-./+%+ 0000 000000 ,-./+%+ 00000 000000 ,-./+%+000000000 ,-./+%+00000 000000 ,-./+%+ 0000 00000000 ,-.1/+%+ 00 0000000000 ,2/+%+ 00000000000 ,2/+%+00000000000 ,2/+%+3,2/+%+0000000000 ,2/+%+ 0000000000 ,2/+%+000000000000 BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 5 o 11 (page numbe no o ci a ion pu poses) ype BOSg _PRLR1 di e by h ee nonsynonymous and by se en synonymous subs i u ions, whe eas he yak hap- lo ype di e s om he Ame ican bison haplo ype BBI_PRLR1 by h ee nonsynonymous subs i u ions. The haplo ype BBI_PRLR1 di e s by i e nonsynonymous and by six synonymous subs i u ions om he BOS_PRLR1. Sheep haplo ypes di e om each o he wi h one o wo subs i u ions, OVIS_PRLR1 being he majo sheep haplo- ype (Figu e 5b). Figu e 5c p esen s he pa simony ne - wo k om he Sus da a. Subs i u ion G534S is he only polymo phism ha occu s wice in he ne wo k. The majo wild boa haplo ype SUS_PRLR4 is sepa a ed om he wo mos di e ged domes ic pig haplo ypes SUS_PRLR2 and SUS_PRLR3 by i e nonsynonymous subs i u ions. Discussion In his s udy we demons a e di e en pa e ns o a ia- ion in PRLR and GHR in uminan s and pigs, sugges ing di e gen e olu ion and selec ion p essu es be o e and a e domes ica ion. The PRLR molecula ana omy shows in e es ing di e - ences om he GHR. The Bo inae PRLR haplo ype ne - wo k is e y di e en om ha o he GHR [10], whe e di e ences wi hin Eu opean b eeds, wi hin A ican b eeds and be ween Eu opean/A ican b eeds we e la ge han be ween di e en species (ca le, yak, Ame ican bison as well Eu opean bison (da a om Eu opean bison ob ained om his s udy)). The PRLR ne wo k i s be e o he known phylogene ic ela ionships be ween species. Wi hin ca le b eeds, Eu opean ca le a e lacking majo p i a e PRLR haplo ypes in con as o he p e ious GHR s udy. Ins ead he wo majo haplo ypes, BOS_PRLR1 and BOS_PRLR2 a e sha ed be ween Eu opean and A ican ca le. Du ing pig domes ica ion in Eu ope (s a ed abou 6000 yea s ago), domes ica ed p ogeny o he local Eu opean wild boa s eplaced in oduced nea eas e n domes ic pigs [15]. Acco ding o he coalescen heo y he mos equen haplo ype in a popula ion le el s udy is he ances al hap- lo ype [16], on he o he hand Wa e son e al [17] a gued ha he highe he mu a ion a e, he less likely i is ha he mos equen allele would be he oldes . Conside ing he popula ion his o y o he pigs, we assume ha he majo wild boa haplo ype SUS_PRLR4, e en hough no he mos equen haplo ype is a good candida e o being he ances al haplo ype o he Eu opean domes ic pig. I ue, subs i u ions leading o es o he domes ic pig Sus_PRLR haplo ypes ha e happened a e he domes ica- ion, as he haplo ypes a e de i ed by a se ies o consecu- i e subs i u ions. The esul is simila wi h he esul ob ained om he MC1R gene ha has an e ec on coa colou in pigs [18]. MC1R alleles/haplo ypes di e by S a is ically in e ed sheep haplo ypes om he PRLR exon 10Figu e 2 S a is ically in e ed sheep haplo ypes om he PRLR exon 10.              !  .    )'+ -$4./+%+ 000 -$4./+%+00 -$4./+%+00000 -$4./+%+00 -$4./+%+000 -$4./+%+ 000 S a is ically in e ed pig haplo ypes om he PRLR exon 10Figu e 3 S a is ically in e ed pig haplo ypes om he PRLR exon 10.                % &  )       +%. .$ .5./+%+ 00 .5./+%+00 .5./+%+0 .5./+%+0 .5./+%+00 BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 6 o 11 (page numbe no o ci a ion pu poses) mo e han one nonsynonymous mu a ion om he wild- ype, implying a long his o y o s ong posi i e selec ion o coa colou a ian s. I is a gued ha coa colou phe- no ypes esul om di ec human selec ion [18]. Th ee domes ic pig haplo ypes di e om he SUS_PRLR4 by wo o mo e nonsynonymous subs i u ions, he la ges di e ence being i e nonsynonymous subs i u ions. The apid accumula ion o mu a ions in he PRLR gene in he pig is bes explained by human in luence. Domes ica ion and ollowing selec ion o ag icul u al pu poses ha e educed b eed e ec i e popula ion sizes o ela i ely small numbe s. In a popula ion wi h low Ne a la ge ac- ion o sligh ly dele e ious mu a ions can each ixa ion due o less e icien pu i ying selec ion [19]. This may p o- ide one explana ion o he obse ed polymo phism in PRLR. Howe e , he GHR gene is monomo phic in he same pig sample, sugges ing di e en selec ion p essu es (selec i e sweep o s ong pu i ying selec ion in GHR, s ong posi i e selec ion o PRLR). A e domes ica ion he ep oduc ion pe o mance o he pig has inc eased se e al olds due o an inc ease in li e s pe yea and pigle s pe li e [20]. This has been made possible by selec ion and making he en i onmen mo e a ou able o he sow o alloca e esou ces o mo e o - sp ing. Many s udies in pigs ha e e ealed signi ican associa ion o PRLR polymo phisms wi h di e en ep o- duc ion ai s (e.g. [11,21,22]). Mouse knockou models o he PRLR ha e con i med he impo ance o he PRL in ep oduc ion ( e iewed by [23]). In addi ion o ep oduc- i e pe o mance, he lis o modes o ac ions o PRLR is long [8] and shows conside able change h ough e e- b a e e olu ion [9]. I a gene has unde gone epea ed al e na ing unc ional adap a ion o ul il one o wo, e en mo e, unc ions, i can lead o accumula ion o many amino acid subs i u- ions wi h sligh ly li le o e all change o unc ion. This phenomenon has been called unc ional swi ching [24] and could hus explain PRLR polymo phism. Based on he s udies done wi h he domes ic animals (e.g. [6,7,11,21,22,25]), PRLR could ul il he c i e ion o unc- ional swi ching. Howe e , he accumula ion o mu a- ions in he domes ic pig PRLR seems oo as o be explained by unc ional swi ching alone. The majo ole o GHR in ol es pos na al g ow h and mamma y unc ion. The lack o a ia ion in he GHR sig- nalling domain in pigs and undomes ica ed uminan s is in iguing. The lack o a ia ion in pigs could be he esul Table 2: P edic ed a ec ion s a us o he amino acid subs i u ions om PRLR gene exon 10 SIFT PolyPhen Species p edic ion sco e MSC n p edic ion PSIC±n Eu opean Ca le (Bos au us) P340T ole a ed 0.35 3.1 30 benign 0.78 33 E378K a ec p o ein unc ion 0.02 3.12 28 benign 0.88 27 A536V ole a ed 0.34 3.1 29 benign 0.41 31 A ican Ca le (Bos indicus) V439M ole a ed 0.07 3.1 29 benign 0.39 31 L497R ole a ed 0.36 3.1 29 benign 0.13 30 Ame ican bison (Bison bison) E384K* ole a ed 0.06 3.08 27 benign 1.48 23 Eu opean bison (Bison bonasus) E384K* ole a ed 0.06 3.05 29 benign 1.50 22 Q394K* ole a ed 0.16 3.05 29 benign 1.21 24 M445V* ole a ed 0.42 3.05 28 benign 1.48 22 Sheep (O is a ies) E387K ole a ed 0.47 3.05 29 benign 0.57 31 A476T ole a ed 0.64 3.05 29 benign 0.10 30 S480R ole a ed 0.62 3.05 29 benign 0.24 31 Domes ic pig (Sus sc o a) L406P ole a ed 0.25 3.16 27 benign 0.65 23 D428A ole a ed 0.54 3.15 29 benign 0.04 28 A461G ole a ed 0.36 3.15 29 benign 1.34 28 K480R ole a ed 0.61 3.15 29 benign 0.29 28 M510L ole a ed 0.64 3.15 29 benign 1.31 28 G534S ole a ed 0.79 3.17 25 benign 0.23 23 G597S ole a ed 0.65 3.15 25 benign 0.35 18 A601V ole a ed 0.23 3.15 25 benign 0.51 19 * analyses we e done using he cy oplasmic pa alone, MSC = median sequence conse a ion, n = numbe o sequences ep esen a his posi ion in he p o ein alignmen , ± PSIC sco e di e ence BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 7 o 11 (page numbe no o ci a ion pu poses) o a s ong selec i e sweep ha has happened be o e domes ica ion and has been main ained in domes ic pigs. On he o he hand, he low di e si y in he wild boa may e lec a na ow gene ic backg ound o he Finnish wild boa popula ion. The possible a ge (causa i e mu a ion) o selec ion a o close o he GHR exon 10 is no known. In con as , he GHR gene cy oplasmic domain in cu en ca le b eeds ha bou s lo s o in e es ing pe sis en poly- mo phism, indica ing i s possible impo ance o ca le being malleable o a i icial selec ion o g ow h and lac- a ion ai s [10]. The PRLR gene on he o he hand shows con as ing e o- lu ion in he s udied species, ha ing mo e subs i u ion polymo phism in pig b eeds and bison species han in ca le. Func ional analysis indica ed possible e ec s o nsSNPs ound in ca le and bison, bu didn' e eal any e ec on he p o ein s uc u e o pig nsSNPs. Howe e , me hods de ining e ec s o SNPs a e only p edic ions, and he combined e ec s o a ious nsSNPs can no be analysed by cu en s a is ical me hods. No known 3D s uc u e o PRLR is a ailable, so mo e de ailed analysis whe he he subs i u ion si es a e in spa ial con ac wi h c i ical esidues emains unknown. Func ional analysis a he molecula le el would be necessa y o sol e he ele- ance o he polymo phisms. Domes ica ion is a cumula i e p ocess ma ked by changes on bo h sides o he mu ualis ic ela ionship. Long-li ing animals like ca le espond slowe o selec ion han o example annual c ops [1]. T ai s ha ha e been he a ge o domes ica ion and subsequen selec ion and he genes ha a ec hem, i.e. so called domes ica ion genes, a e o special in e es e en hough he success o iden i y hese genes om li es ock has been low [1,4]. Genes iden i ied by QTL s udies a e good candida es o genes impo an o domes ica ion success [1]. Conclusion QTL candida e genes GHR and PRLR ha e a ying oles in g ow h, lac a ion and ep oduc ion in domes ic species and may ha e esponded di e en ly in di e en species and b eeds wi hin species o di e en selec ion p essu es. We show he e ha hese genes ha e di e en e olu iona y his o y among a iodac yls. Mos likely bo h genes ha e been hea ily in luenced by a i icial selec ion du ing and a e domes ica ion. We ound an unan icipa ed amoun o nonsynonymous a ia ion accumula ed o pe sis ed in hese genes in a iodac yl species du ing he sho pe iod o domes ica ion (8000 – 10 000 yea s) and subsequen a i icial selec ion. The di e ences be ween species indi- ca e possible e ec s o selec i e sweeps be o e domes ica- ion (GHR in pigs) o be o e species di e gence (GHR in Bison), di ec ional (a i icial) selec ion (PRLR in pigs) o unc ional swi ching (GHR in ca le, PRLR). Sliding window p esen a ion o he nucleo ide di e gence along PRLR exon 10 among 18 di e en speciesFigu e 4 Sliding window p esen a ion o he nucleo ide di e gence along PRLR exon 10 among 18 di e en species. B a = ca le, Bbi = Ame ican bison, Bbo = Eu opean bison, O is = sheep, Sus = pig. The posi ions o he de ec ed SNPs a e shown on he X-axis. BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 8 o 11 (page numbe no o ci a ion pu poses) Me hods Ma e ials We in es iga ed 216 indi iduals om 12 di e en Eu o- pean Bos au us b eeds (comme cial dai y b eeds: Finnish Ay shi e, Finnish Hols ein-F iesian; na i e Finnish dai y b eeds: Wes e n Finnca le, No he n Finnca le, Eas e n Finnca le, Russian b eeds: Kholmogo , Yaku Ca le, Bes- uzhe , and Belo ussian Red, and Sou heas Eu opean b eeds: Busa, Podolian, and Uk ainian G ey). In addi ion, we s udied 22 samples om 3 di e en A ican b eeds om E hiopia wi h unknown Bos indicus backg ound (Ba ka; mo phological zebu- ype, Raya; mo phological sanga- ype, and Foge a; zebu-sanga in e media e). Ca le samples o igina ed om 6 di e en coun ies (Finland, Russia, Uk aine, Byelo ussia, Se bia, and E hiopia). As e - e ences we s udied 3 Ame ican bison indi iduals (Bison bison) om one b eeding s ock, 6 Eu opean bison (Bison bonasus) samples om Poland, and 4 yak (Bos g unniens) samples om Russia. O he s udied species included 14 sheep (O is a ies) ep esen ing 6 di e en b eeds om Russia, Poland and Finland (Romano b eed, W zosowka b eed, Dages an local, Andi, sheep om Komi illage, Finnsheep, and Ålandsheep), 18 pigs (Sus sc o a) ep e- sen ing La ge Whi e, Land ace, Hampshi e and Du oc b eeds, and 9 wild boa samples (Sus sc o a) om h ee di e en Finnish wild boa a ms. Indi iduals om he Pa simony ne wo k econs uc ionsFigu e 5 Pa simony ne wo k econs uc ions. a) Bo inae haplo ype ne wo k. Di e en colou s ep esen di e en ca le b eed g oups and/o species. b) O is haplo ype ne wo k c) Sus haplo ype ne wo k. The size o he pie is p opo ional o he e- quency o he haplo ype o he b eed g oup in ques ion and he size o he node a ea is p opo ional o he o al equency o he haplo ype in he whole popula ion. Small black ci cles ep esen he hypo he ical haplo ypes no p esen in his s udy. BOS = ca le, Bbi = Ame ican bison, Bbo = Eu opean bison, BOSg = yak, SUS = pig, OVIS = sheep. BMC E olu iona y Biology 2009, 9:172 h p://www.biomedcen al.com/1471-2148/9/172 Page 9 o 11 (page numbe no o ci a ion pu poses) same b eed/species we e sampled o be as un ela ed as possible. Eu opean bison is an endange ed species and all cu en animals (app oxima ely 3200 indi iduals) descend om 12 ounde animals [26]. All he s udied species belong o he same o de , A iodac yla. Pig belongs o he amily Suidae, while all o he species belong o he amily Bo idae. Ca le and yak a e om he genus Bos, while Ame ican and Eu opean bison belong o genus Bison. Gene ic analysis Genomic DNA was ex ac ed om blood o semen sam- ples using sal ing ou p ocedu e [27]. The p olac in ecep- o gene exon 10 was ampli ied in wo agmen s wi h he same p ime se s om ca le, yak, and bison samples. Exon 10 om he pig and sheep samples was ampli ied in one agmen . P ime s we e designed wi h he P ime 3 h p:// odo.wi.mi .edu/ using e e ence sequences [Gen- Bank: NM_001039726 and DQ458765.1]. The PRLR exon 10 sequence o he uminan comp ises o 891 bp (297 aa), which was sequenced en i ely, and ha o he pig is 1023 bp (341 aa) long, o which we sequenced 750 bp (250 aa). In addi ion, Eu opean bison and wild boa GHR exon 10 we e analysed using same me hods and p ime s as desc ibed in [10]. P ime sequences a e a aila- ble om TI-T upon eques . In polyme ase chain eac ion (PCR), 50 ng o genomic DNA was used in 30 Pl olume o s anda d DYNAZyme II (Finnzymes, Finland) PCR eac ion mix. PCR p oduc s 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 wi h DYEnamic ET Te mina o Ki (GE Heal hca e Li e sci- ences, UK). The sequencing p oduc s we e pu i ied wi h e hanol p ecipi a ion and sepa a ed on MegaBACE 1000 (Ame sham Biosciences, UK). Each agmen was sequenced on bo h s ands and same p ime s we e used o he sequencing as o he agmen ampli ica ion. Sequence da a was base-called wi h Cima on 3.12 base- calle in p og am MegaBACE Sequence analyze . 3.0.0111.1603 (Ame sham Biosciences, UK). All sequences we e e i ied by isual inspec ion o ch oma o- g ams. Sequenche 4.6 (Gene Codes Co po a ion) was used o align and co ec sequences. S a is ical analysis Haplo ypes we e econs uc ed using he Bayesian haplo- ype econs uc ion me hod o single nucleo ide poly- mo phisms (SNP) om popula ion geno ype da a u ilizing he p og am PHASE .2.1.1 [28]. Recons uc- ions we e done using 10000 i e a ions, 1 hinning in e - al and 1000 as bu n-in pe iod assuming a s epwise mu a ion model. Animals ha go haplo ype pai p oba- bili y lowe han 0.95 we e ei he cloned o disca ded om u he analysis. Cloning was done wi h Ze o Blun ® TOPO® PCR Cloning Ki (In i ogen, USA) and Phusion™ High-Fideli y DNA Polyme ase (Finnzymes, Finland) ol- lowing manu ac u e 's ins uc ions. Fi e o mo e clones pe indi idual we e sequenced as desc ibed abo e using uni e sal M13-p ime s. Haplo ypes we e decided acco d- ing o he esul ob ained om he cloning. We es ima ed nucleo ide di e si y, S (pi, [29]) and Wa e - son's he a es ima o , TW [30] along he exon 10 and made a sliding window plo om hese wo pa ame e s among 18 e eb a e species. Pi and he a we e calcula ed o 10 bp windows placed a 5 bp in e als using he ollowing GenBank h p://www.ncbi.nlm.nih.go sequences: NM_204854.1 (chicken), L76587.1 ( u key), XM_001508575.1 (pla ypus), NM_001039726.1 (cow), NM_000949.2 (human), NM_011169.4 (mouse), NM_001034111.1 ( a ), NM_001082231.1 (Eu opean abbi ), XM_001500104.1 (ho se), XM_536502.2 (dog), NM_001001868.1 (pig), NM_001085736.1 and NM_001085616.1 (Zenopus lae is), NM_001124599.1 ( ainbow ou ), DQ508436.1 (salmon), XM_001921376.1 (Danio e io), NM_001078625.1 ( ak- i ugu), XM_001149377.1 (chimpanzee), and XM_001091532.1 ( hesus monkey). We calcula ed haplo ype di e si y (Hd), nucleo ide di e - si y (pi) and Wa e son's he a es ima o o he s udied species sepa a ely using haplo ype sequences ob ained. Pi is based on he a e age numbe o nucleo ide di e ences be ween he sequences and he a is based on he o al numbe o seg ega ing si es in he sequence. To es ima e he e ec o selec ion, we calcula ed Tajima's D [31], and Fu and Li's D* and F* [32] o each species/subspecies sepa a ely. Tajima's D es compa es he di e ences be ween he numbe o seg ega ing si es and he a e age numbe o pai wise di e ences [31]. Unde neu ali y, Tajima's D alue is assumed o be ze o, unde posi i e selec ion he e is an excess o a e polymo phisms and Tajima's D alue is nega i e. Nega i e D alues can also be due o popula ion expansion. I he e is balancing selec- ion in e media e equency gene ic a ian s a e kep and Tajima's D alue is posi i e. The same holds o he Fu and Li's D* and F* es s. Con idence in e als o he Tajima's D and Fu and Li's D* and F* alues we e ob ained by gen- e a ing 1000 independen coalescen simula ions assum- ing no ecombina ion. S a is ical analysis package DnaSP 4.50.3 [33] was used o calcula e Hd, Pi, Wa e son's he a es ima o , Tajimas's D alues and Fu and Li's D* and F*. The impac o amino acid a ian s on p o ein s uc u e ia analysis o mul iple sequence alignmen s was done wi h SIFT [13] and PolyPhen so wa e [14]. SIFT uses sequence homology o p edic whe he an amino acid subs i u ion will a ec p o ein unc ion and hence, po en ially al e pheno ype. I gi es no malized p obabili y sco e alue