A region on chicken chromosome 2 affects both egg white thinning and egg weight

Gene . Sel. E ol. 37 (2005) 563–577 563 c INRA, EDP Sciences, 2005 DOI: 10.1051/gse:2005016 O iginal a icle A egion on chicken ch omosome 2 affec s bo h egg whi e hinning and egg weigh Me i Ha∗,Ma iaT-Ha, Di k-Jan  Kb, Anneli Va,AskoM ¨ -Ta, Johanna Va aAnimal P oduc ion Resea ch, MTT Ag i ood Resea ch Finland, 31600 Jokioinen, Finland bRoslin Ins i u e, Roslin, Midlo hian EH25 9PS, UK (Recei ed 28 Feb ua y 2005; accep ed 2 May 2005) Abs ac – We desc ibe he esul s om gene ic dissec ion o a QTL egion on chicken ch o- mosome 2, shown o affec egg weigh and quali y in an ea lie genome scan o an F2in e c oss be ween wo di e gen egg laye lines. As he 90% con idence in e als o he de ec ed QTL co e ed ens o cen iMo gans, new analyses we e needed. The da ase s we e eanalysed wi h dense ma ke in e als o cha ac e ise he QTL egion. Analysis o a candida e gene om he o iginal QTL egion, imen in, did no suppo i s ole in con olling egg whi e hinning. E en a e eanalysis wi h addi ional se en ma ke s in he QTL a ea, he 90% con idence in e als emained la ge o e en inc eased, sugges ing he p esence o mul iple linked QTL o he ai s. A g id sea ch i ing wo QTL on ch omosome 2 o each ai sugges ed ha he e a e wo dis- inc QTL a eas affec ing egg whi e hinning in bo h p oduc ion pe iods and egg weigh in he la e p oduc ion pe iod. The esul s indica e possible pleio opic effec so someo heQTLon egg quali y and egg weigh . Howe e , i was no possible o make a dis inc ion be ween close linkage e sus pleio opic effec s. chicken /ch omosome 2 /egg quali y /egg weigh /QTL mapping 1. INTRODUCTION Du ing he las en yea s, many QTL ha e been de ec ed in poul- y affec ing p oduc ion ai s including g ow h a e, ca cass ai s and eed efficiency [4, 6, 10, 13, 15, 27, 32, 37, 38], as well as disease esis- ance [36, 42–44, 46]. Mos o hese s udies we e done on expe imen al chicken popula ions om c osses in ol ing b oile s o exo ic lines. Fa less a en ion ∗Co esponding au ho : me i.honka ukia@m . i The nucleo ide sequence da a epo ed in his pape ha e been submi ed o GenBank Submis- sion nucleo ide da abase (Accession Numbe s AY694786–AY694792). A icle published by EDP Sciences and a ailable a h p://www.edpsciences.o g/gse o h p://dx.doi.o g/10.1051/gse:2005016 564 M. Honka ukia e al. has been paid o egg quali y ai s such as egg whi e hinning o shell quali y, al hough hese a e also economically impo an [40]. In a ecen s udy Sasaki e al. [24] epo ed QTL affec ing egg quali y ai s (egg shell s eng h, shell colou , shape o egg), howe e egg whi e quali y was no included. One o he majo con ibu o s o he in e io quali y o eggs is he egg whi e. Egg whi e hinning is a sign o quali y loss ha leads o s aleness. F esh and i m egg whi e es ic s mic obial g ow h because egg whi e p o eins ha e an i- mic obial p ope ies [21], while hin and “ unny” egg whi e exposes he egg o mic obial in ec ions due o al e ed s uc u e and biochemical composi ion. Thin egg whi e also has an un a ou able impac on he ha chabili y o chicken emb yos [2]. F esh egg albumen consis s o ou majo pa s: wo hick and wo hin whi es in addi ion o chalazae. The majo po ion (app oxima ely hal ) o he albumen consis s o he ou e hick whi e. The hick whi e is a i m jelly-like gel ha con ains mic oscopic ib ils. The chalazae and he hick egg whi e a e equi ed o keep he yolk in he cen e o he egg. I is impo an ha any con ac be ween he yolk and shell memb anes is a oided in o de o p e en mic obial g ow h in he yolk. A la ge p opo ion o hick egg whi e is also de- si able in he ood indus y o oaming and hea coagula ion o baking. When he egg is s o ed, he albumen becomes mo e luid and less gela inous [17]. Howe e , his a ies g ea ly acco ding o b eed and heal h condi ions o he hens. Albumen also hins du ing incuba ion. P obably he dec ease in iscos- i y is necessa y o he apid diffusion o albumen p o ein o he emb yo [2]. Egg whi e hinning is egula ed by e y complex p ocesses [3] and can be ega ded as a quan i a i e ai . The he i abili y (h2) es ima es o albu- men quali y ange om e y low o high [41], an a e age es ima e being 0.38 [19, 20, 28]. The gene ic co ela ion be ween egg whi e hinning and egg p oduc ion (egg numbe ) is sligh ly nega i e [19]. The gene ic co ela ion be- ween egg size and albumen quali y has been nega i e in se e al s udies ci ed by Washbu n [41]. As a esul , selec ion o inc eased egg p oduc ion may esul in dec eased egg whi e quali y. In some si ua ions (e.g. when he espon- sible genes a e closely linked), i could he e o e be ad an ageous o be able o di ec ly selec indi iduals by geno ype o linked ma ke s. We ha e p e iously de ec ed QTL egions affec ing egg whi e hinning on ch omosomes 2, 4 and 8 [35]. The QTL egion on ch omosome 2 ha bou - ing genome-wide signi ican QTL effec s on bo h egg whi e hinning and egg weigh was chosen o u he s udy. The 90% con idence in e al a ea coin- cides wi h he loca ion o se e al pu a i e egg quali y candida e genes, includ- ing imen in among o he s [25]. Vimen in belongs o a amily o in e media e Gene ic dissec ion o egg quali y QTL 565 ilamen s, which a e impo an in main aining he mechanical in eg i y o he cell [23,45]. Mo eo e , he in e media e ilamen s make s iffgel-like s uc u es h ough polyme isa ion [30]. In he p esen s udy, we desc ibe he gene ic dissec ion o he ch omosome wo QTL egion. Mul iple ma ke eg ession, including g id sea ch o mul iple QTL, was used oge he wi h dense ma ke in e als. 2. MATERIALS AND METHODS 2.1. Mapping popula ion and pheno ypic obse a ions The mapping popula ion was based on a ecip ocal in e c oss o wo pa en al lines, he Whi e Legho n (WL) and Rhode Island Red (RIR). The pa en gen- e a ion consis ed o ou animals pe line. The F2popula ion o 305 indi id- uals was achie ed by ma ing eigh F1males wi h 32 F1 emales [35]. The wo pa en al chicken lines di e ged mainly o egg quali y. The RIR line eggs we e sligh ly ligh e (61.5 g) han hose o he WL (63.9 g), and he HU sco - ing o egg whi e heigh in he RIR line was lowe (53.0 HU) han in he WL line (91.0 HU). The p oduc ion pe iod was di ided in o wo pa s. The i s pa includes a pe iod om 18 o 40 weeks o age, du ing which ime egg p oduc ion eaches he highes po en ial and op imal egg weigh . Du ing he second pe iod om 41 o 60 weeks o age he egg whi e quali y slowly dec eases, whe eas egg weigh inc eases [8, 31]. Albumen heigh was sco ed in Haugh uni s (HU). HU is a unc ion o he heigh o he hick albumen and he weigh o he egg co ec ed o a cons an 56.7 g egg [11]. The HU sco ing was done h ee imes. Fi s , o eggs col- lec ed du ing 36 o 39 weeks (HU40), secondly o eggs collec ed du ing 57 o 59 weeks o age (HU60). Fo each hen, one egg was collec ed pe week; he HU alues used in he QTL analysis we e he a e ages wi hin each obse a ion pe iod. Fo he hi d ime, he effec o s o age was s udied om eggs laid a he age o 60 weeks and he ea e s o ed o 28 days ( he leng h o he sales pe iod on he ma ke ) a +8◦C (s o ed HU) o de ec any possible gene ic e - ec on he a e o albumen quali y decline. In he F2mapping popula ion, he mean o s o ed HU was 57.6 wi h a s anda d de ia ion o 10.0. Egg weigh was eco ded as he mean o weekly measu emen s (1 egg/week) du ing wo pe iods: ea ly p oduc ion, EWa (be ween 18 and 40 weeks o age) and la e p oduc ion, EWb (be ween 41 and 60 weeks o age). The pheno ypic 566 M. Honka ukia e al. da a we e p e-adjus ed o ha ch effec . No o he sys ema ic en i onmen al effec s we e obse ed o adjus ed. 2.2. Ma ke geno yping The i s genome scan inco po a ed geno ypes o 13 ma ke s spanning 351 cM o ch omosome 2 [35]. The new analysis was done wi h six addi- ional mic osa elli e ma ke s (MCW0220, LEI0117, MCW0243, MCW0206E, LEI0089 and MCW0264) and an SNP ma ke (VIMin 4) wi hin he QTL egion. The addi ional mic osa elli es we e chosen om Comp on (C), Eas Lansing (EL) and Wageningen (WAU) gene ic maps (h ps://acedb.asg.wu .nl). In addi ion, se e al ma ke s we e es ed bu excluded on he g ounds o be- ing unin o ma i e (LEI0247, MCW0142, ADL309, MCW0131, ADL0197, ADL0185, LEI0086 and MCW0184).Mic osa elli e ma ke geno yping and analysis we e done using s anda d p o ocols [35]. MegaBACE Gene ic P o- ile (Ame sham Biosciences) so wa e was used o de e mine he geno ypes o he mic osa elli e ma ke s. Sequence in o ma ion was a ailable om chicken imen in [45], and i was analysed o a ia ion wi hin he mapping popula ion. VIMin 4 is a single nucleo ide polymo phism wi hin he imen in gene, de ec ed by sequencing ou indi iduals o he mapping popula ion. Sequencing p ime s we e de- signed based on in o ma ion o he DNA sequence o he chicken imen in gene, GenBankTM/EMBL accession numbe J02759 (gi: 212864–212866). A ew animals ep esen ing ex eme HU alues we e selec ed o be sequenced. Se e al polymo phic si es we e iden i ied in imen in, mainly in he in ons (Tab. I). One o he polymo phisms, VIMin 4 (an A o G change wi hin he in on be ween exons 4 and 5) was used as a ma ke in linkage analysis. The VIMin 4 was sco ed using minisequencing Snupe Geno yping (Ame - sham Biosciences). An ampli ied PCR p oduc o he egion ( o wa d p ime : 5’– GTG TCC TCT TCG AGT GAG TG; and e e se: 5’– GAA AAT CCA AAA CAT GTA) was used as a empla e in a he mally cycled minisequenc- ing eac ion wi h dye-labelled e mina o s and sequence speci ic p ime ending one base be o e he polymo phic base (5’– CCT GGG TCT AGA AGT AGT CC). Excess deoxynucleo ides and p ime s we e emo ed wi h speci ic enzy- ma ic ea men o he eac ion p oduc be o e minisequencing (ExoSap-IT TM Ame sham Biosciences). Uninco po a ed dye-labelled e mina o s we e subse- quen ly emo ed by il a ion (Au oSeqTM96 pla es, Ame sham Biosciences). Reac ion p oduc s we e de ec ed on a capilla y DNA elec opho esis ins u- men (MegaBACE, Ame sham Biosciences). Analysis was based on allele Gene ic dissec ion o egg quali y QTL 567 Table I. Loca ion o polymo phisms and de ec ed nucleo ide a ian s in he imen in gene. Posi ion # n o #exon /in on Polymo phisms genebank sequence 255 o 212869 5’ lanking A /G 259 o 212869 5’ lanking A /T 406 o 212869 5’ lanking A /G 613 o 212869 5’ lanking C /T 643 o 212869 5’ lanking C /T 762 o 212869 5’ lanking A /C 257 o 323864 5’ lanking A /G 520 o 212864 5’ lanking A /G 1207 o 212864 exon 1 A /G 21 o 212865 exon 3 C /G 145 o 212865 in on 3 A /G 107 o 2128661in on 4 A /G 328 o 212866 in on 5 T /G 381 o 212866 in on 5 T /G 389 o 212866 in on 5 C /T 431 o 212866 in on 5 C /T 818 o 212866 in on 5 C /T 892 o 212866 in on 5 A /C 1173 o 212866 in on 6 A /C 1796 o 212866 in on 8 C /T 1875 o 212866 in on 8 A /G 2139 o 212866 in on 8 A /G 2169 o 212866 in on 8 C /T 1VIMin 4, he SNP included in he linkage analysis. signal in ensi y and ela i e allele mobili y using MegaBACE SNP P o ile , e sion 1.0 (Molecula Dynamics 2001). 2.3. Linkage analysis A linkage map o he 20 ma ke s on chicken ch omosome 2 was cons uc ed using CRIMAP, e sion 2.4 [9]. All pai wise combina ions o ma ke s we e i s checked using he “ wo-poin ” op ion. The o de o ma ke s wi hin he 568 M. Honka ukia e al. linkage g oup was de e mined wi h op ion BUILD. The supe io i y o he cu - en o de o adjacen ma ke s was compa ed wi h possible op ional o de s us- ing FLIPS. Op ion CHROMPIC was used o de ec geno yping e o s (double o iple ecombinan s). 2.4. Reg ession analysis QTL Exp ess, a web-based applica ion (h p://q l.cap.ed.ac.uk) [26] o lin- ea eg ession, was used o QTL analyses, as well as a simila cus om-made p og am o pe o m addi ional pe mu a ions o he da a, and boo s apping o ob ain he h eshold alues o s a is ical es and he 90% con idence in e als. Analyses we e done by i ing ei he one o wo QTL simul aneously o each ai (HU40, HU60, s o ed HU o EWa, EWb) on ch omosome 2. An empi ical pe mu a ion es (10 000 i e a ions) was applied o de e - mine signi icance h esholds o 5% and 1% ch omosome wide signi icance le els. Con idence in e als o QTL posi ions we e de e mined by boo s ap- ping [39]. The 90% cu -offpoin s o so ed bes es s a is ics om 10 000 boo - s ap eplica es we e calcula ed using he cus om-made p og am and we e used o de ine he bounda ies o he con idence in e al [5]. Fo he wo QTL model, an F- es was used o de e mine whe he he bes QTL pai explained signi ican ly mo e o he a ia ion han he bes single QTL om he pai . 3. RESULTS 3.1. QTL mapping The a e age in o ma ion con en was >0.6 along he ch omosome. This was a clea imp o emen compa ed wi h he p e ious s udy [35]. The new analy- ses wi h 20 ma ke s showed a signi ican QTL affec ing HU40, wi h he bes posi ion a 137 cM be ween ma ke s MCW0206E (114 cM) and ADL0217 (152 cM), hus alling in he p e iously es ablished QTL a ea (Tab. II). The effec o he RIR allele was –3.73 HU (±0.80), while he dominance effec was –1.74 HU (±1.51). The de ec ed QTL explained 6.7% o he pheno ypic a iance. The 90% con idence in e al o he QTL loca ion was b oadened om he p e ious 58 cM o 64 cM despi e he dense ma ke map. Auxilia y analyses, i ing wo QTL simul aneously, indica ed he exis ence o wo dis- inc QTL a eas affec ing HU40 a 141 cM (he ea e named HU QTL-1) and 54 cM (HU QTL-2) (Tab. III). Gene ic dissec ion o egg quali y QTL 569 Table II. QTL o egg whi e quali y (HU40, HU60) and la e egg weigh (EWb) by one QTL model in ch omosome 2. The effec was calcula ed as he effec o he allele om he pa en al RIR line. T ai F- a io1Posi ion (cM) Addi i e Dominance 90% C.I o effec effec QTL posi ion2 (s.e) (s.e) HU40 11.34 137 –3.73 (±0.80) –1.74 (±1.51) 97–161 (p <0.01) HU60 7.06 139 –5.50 (±1.48) ns 128–150 (p <0.01) EWa 4.05 137 ns ns ns (p >0.05) EWb 6.54 12 –0.45 (±0.60) –3.80 (±1.09) 0–48 (p <0.05) 1Ch omosome-wide signi icance shown in pa en heses. 2Empi ical con idence in e als based on F a ios (boo s apped 10 000 imes, cus om-made p og am). Table III. QTL o egg whi e quali y (HU40, HU60) and la e egg weigh (EWb) om he wo QTL model. The effec was calcula ed as he effec o he allele om he pa en al RIR line. T ai F a io F- a io Posi ions Add. effec Dom. effec Add. effec Dom. effec 2 s. 0QTL 2 s. 1 QTL o QTL o QTL 1 o QTL 1 o QTL 2 o QTL 2 HU40 7.89 4.13 54 (QTL2) and –3.38 ns –1.49 2.26 (p <0.001)1(p <0.025)1141 (QTL1) (±0.82) (±0.70) (±1.06) HU60 6.58 5.82 32 (QTL2) and –5.90 –3.11 ns 6.70 (p <0.005)1(p <0.005)1140 (QTL1) (±1.51) (±2.80) (±2.03) EWb 6.37 5.98 12 (QTL2) and –1.70 1.08 ns –4.09 (p <0.005)1(p <0.005)1(QTL1) 83 (±0.54) (±0.79) (±1.10) 1nominal p-le el. In he p e ious genome scan no s ong e idence o a QTL affec ing he egg whi e hinning in he la e laying pe iod (HU60) was ound. Howe e , eanalysis wi h he dense map e ealed a signi ican QTL effec on HU60, he mos p obable loca ion being 2 cM dis an om he one o HU40 (Tab. II). 570 M. Honka ukia e al. The addi i e effec on HU60 was highe (–5.50 ±1.48) han i was on HU40 and also he 90% con idence in e al was na owe (22 cM). The analyses o i ing wo QTL e ealed a second QTL a ea a 32 cM. The e he addi i e effec was no signi ican , bu he dominance effec was ma kedly highe han he dominance affec ingHU40a QTL-2(6.70±2.03 e sus 2.26 ±1.06) (Tab. III). A bo h QTL loca ions, he RIR allele had a dec easing addi i e effec on HU sco es in bo h p oduc ion pe iods, which is cong uen wi h egg whi e quali y ea u es o he RIR line. A he HU QTL-2, he dominance effec was ai ly high and con a y o he addi i e effec . In he i s genome scan, no QTL was iden i ied o affec Haugh alues in s o ed eggs. Nei he did eanalysis e eal any signi ican associa ion be ween ma ke polymo phism and egg whi e hinning a ia ion a e s o age. A genome-wide signi ican QTL affec ing egg weigh in he la e p oduc- ion pe iod was p e iously ound on ch omosome 2. The mos likely posi ion (a 65 cM-wide 90% con idence in e al) o he QTL loca ed wi hin a la ge gap be ween he lanking ma ke s MCW0247 and ADL0217. The ma ke in- o ma ion con en a ha posi ion was low (<0.3). In he e ined s udy, he imp o emen o map esolu ion placed he QTL a 12 cM; wi h a signi ican dominance effec (–3.80 ±1.09) (Tab. II). Fi ing wo QTLs indica ed ha he e was mo e han one QTL affec ing he egg weigh in he la e p oduc ion pe iod. The speci ied loca ions o he wo QTLs appea ed o be on bo h sides o he o me ma ke gap a ea. The effec o EWb QTL 2 (posi ion o 12 cM) was dominan (–4.09 ±1.10) and EWb QTL 1 (posi ion o 83 cM) weakly addi i e (–1.70 ±0.54). A bo h posi ions he RIR allele had a nega i e effec on egg weigh . No signi ican effec was ound o egg weigh in he high p oduc ion pe iod (EWa). Howe e , he es s a is ic eached i s highes alue close o HU QTL-1 a ea (Tab. II). 3.2. Sequencing o imen in Almos he en i e chicken imen in gene was sequenced om wo hens wi h high and wo hens wi h low Haugh alues o iden i y possible a ia ion as- socia ed wi h egg whi e quali y. In he coding sequence, wo polymo phisms we e iden i ied be ween good and poo quali y hens. Se e al polymo phisms we e de ec ed in he 5’- lanking a ea and he in ons (Tab. I). One polymo - phism (VIMin 4) was included in he linkage analysis as an SNP ma ke , which was s udied in he en i e F2mapping popula ion. The linkage analysis Gene ic dissec ion o egg quali y QTL 571 Figu e 1. HU40, HU60 and EWb QTL p o iles om single QTL analysis ex- p essed as F a ios. The signi icance o he 5 % ch omosome-wide le el is ma ked as a do ed line. Ma ke posi ions a e indica ed wi h an a ow in he ollowing o - de : ADL0228, MCW0082, ADL0190, MCW0220, LEI0117, VIMin 4, MCW0247, MCW0243, MCW206E, ADL0217, MCW0065, LEI0089, ADL0197, MCW0039, MCW0041, ADL0267, ADL0236, MCW0264, MCW0056, MCW0166. The SNP ma ke VIMin 4 is ma ked wi h a squa ed a ow. placed imen in a 78 cM. Rega ding he o he polymo phisms, no conco - dan sequence diffe ences be ween pheno ypic g oups (low o high HU) we e iden i ied. 4. DISCUSSION Two QTL a eas (HU QTL-1 and HU QTL-2) affec ing egg whi e hinning we e de ec ed on ch omosome 2 wi hin a 100 cM egion (Fig. 1). Wi hin he wo a eas, he mos p obable loca ions o he QTL affec ing quali y a high (HU40) o la e (HU60) p oduc ion pe iods coincide qui e closely (a QTL-1 HU40 a 137 cM e sus HU60 a 139 cM and a QTL-2 HU40 a 54 cM e sus HU60 a 32 cM) sugges ing ha he e is a single QTL a bo h loca ions affec - ing each egg whi e ai s (HU40, HU60) ai ly equally. The e o e i seems ha he gene ic con ibu ion o bo h ai s a e a leas in pa equal. The RIR allele effec a bo h loca ions o bo h p oduc ion pe iods is nega i e, as expec ed om he line cha ac e is ics. Addi ional suppo o QTL affec ing egg weigh in he la e p oduc ion pe- iod was gained: wo QTL we e indica ed using he dense map. The new po- si ions loca ed on bo h sides o he o me ma ke gap. The effec o he RIR allele on egg weigh was also nega i e.