A

2vomtAp©

K

1

inetpmti

f

0

Small Ruminant Research 67 (2007) 279–284

A TG-repeat polymorphism in the 5′-noncoding region ofthe goat growth hormone receptor gene and search for its

association with milk production traits

Andrzej Maj ∗, Małgorzata Korczak,Emilia Bagnicka, Lech Zwierzchowski, Mariusz Pierzchała

Institute of Genetics and Animal Breeding, Polish Academy of Sciences, Jastrzebiec, 05-552 Wolka Kosowska, Poland

Received 30 June 2004; received in revised form 10 August 2005; accepted 25 November 2005Available online 15 February 2006

bstract

A variable TG-repeat polymorphism was found in the goat growth hormone receptor (GHR) gene 5′-noncoding region. In total35 goats belonging to two dairy breeds were genotyped, 10 alleles of the GHR gene were detected. The length of the repeat regionariants (alleles) was from 305 to 346 bp. Among the 10 alleles identified, eight occurred in homozygous systems. The frequencyf the homozygous genotypes was 0.25 for the Polish White Improved breed and 0.28 for the Polish Fawn Improved breed. The

ean heterozygosity (H) and polymorphic content coefficients (PIC) at this locus were similar for both breeds. The association of

he TG-repeat variants was studied with milk production traits in 157 goats using REML method with repeatability, multi-traitsnimal Test-Day Model. No associations were found with dairy traits—milk yield and content of the major milk components (fat,rotein, and lactose). Also, no effect was shown of the GHR genotype on the somatic cell count (SCC).

2005 Elsevier B.V. All rights reserved.

ellites;

eywords: Growth hormone receptor; Gene polymorphism; Microsat

. Introduction

The biological effects of growth hormone (GH)nvolve a variety of tissues and the metabolism of allutrient classes: carbohydrates, lipids, proteins, and min-rals. In farm ruminants, these coordinated changes inissue metabolism alter nutrient partitioning and thuslay a key role in increasing growth performance and

ilk yield (Etherton and Bauman, 1998). Therefore,

here is a great interest in using growth hormone tomprove production in farm animals. Moreover, the gene

∗ Corresponding author. Tel.: +48 22 756 17 11;ax: +48 22 756 16 99.

E-mail address: A.May@ighz.pl (A. Maj).

921-4488/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.smallrumres.2005.11.007

Milk traits; Goat

encoding GH and other genes related to the so called“somatotropic axis” are considered promising candidatemarkers for selection purposes (Parmentier et al., 1999).

Growth hormone actions on target cells depend onGH receptor (GHR) (Burton et al., 1994). GH bind-ing to GHR causes its dimerization, activation of theGHR-associated JAK2 tyrosine kinase, and tyrosyl phos-phorylation of both JAK2 and GHR (Zhu et al., 2001).These events activate a variety of signalling molecules,including MAP kinases, insulin receptor substrates,phosphatidylinositol 3′-phosphate kinase, diacylglyc-

erol, protein kinase C, intracellular calcium, and STATtranscription factors.

The GH receptor is a member of the cytokine/hemato-poietin superfamily of receptors. The gene coding for

ant Re

280 A. Maj et al. / Small Rumin

GHR of most mammalian species consists of nine exons(from 2 to 10) in the translated part and of a long5′-noncoding region that includes several alternativeuntranslated exons, of which only exons 1A, 1B, and 1Chave been studied in detail in bovine GHR gene (Jiangand Lucy, 2001). Distinct promoters regulate transcrip-tion from the alternative exons. The P1 promoter whichregulates growth hormone receptor expression in liver,is associated with exon 1A in cattle and sheep (Jiang etal., 1999).

From recent publications it is known that some pro-ductive traits of cattle, e.g. milk yield and compositionare associated with polymorphism of the GHR gene(Aggrey et al., 1999; Falaki et al., 1996; Blott et al.,2003). Lucy et al. (1998) found a length polymorphismin a TG-repeat (microsatellite) in the P1 promoter located86 bp upstream from the start site of exon 1A in thebovine GHR gene. They found that an 11-TG-repeatallele commonly occurred in Bos indicus cattle whilealleles with 16–20 consecutive TGs are most common intaurine breeds. However, the shorter 11-TG-repeat allelecan be found at low frequency among European cattle,e.g. in Aberdeen Angus. The goat GHR gene has notbeen investigated yet. The objective of this study wasto search for polymorphic microsatellite repeats withinthe 5′-region of the goat GHR gene and to verify possi-ble association between this putative polymorphism andproductive traits.

2. Materials and methods

2.1. Animals

Studies of the GHR gene TG-repeat polymorphismwere conducted on 115 dairy Polish White Improved(PWI) and 120 Polish Fawn Improved (PFI) goats (does,bucks and kids). For studying associations between theGHR polymorphism and milk production traits data for157 dairy goats being from 1st to 6th lactation wereused. Goats of the PWI breed were mated with PWIand Saanen bucks, and goats of the PFI breed with PFIand Alpine bucks. The goats were maintained in threeherds. The goats were kept in a loose barn with outsiderun (except winter). During the test period the animalswere fed according to the INRA-system (Jarrige, 1988);water was available ad libitum.

An authorized veterinarian collected blood samplesfrom jugular vein to tubes containing K3EDTA. DNAfor GHR gene sequencing and genotyping was isolatedfrom blood by the method of Kanai et al. (1994) andamplified by PCR.

search 67 (2007) 279–284

2.2. Sequencing of the caprine GHR gene 5′-region

Basing on the available sequences of the bovine(GenBank U15731) and ovine (O’Mahoney et al.,1994) GHR genes, and using the Primer3 soft-ware (www.genome.wi.mit.edu), primers were designedaimed at PCR amplification of overlapping fragmentsof the caprine GHR gene 5′-region (Table 1). Poly-merase chain reactions were performed using a PCR-mixwith: primers at 5.0 pmol/�l, 1 U Taq polymerase (Pol-gen, Łodz, Poland), 1 �l Taq polymerase buffer, fourdNTPs, each at a final concentration of 0.2 mM, 100 ngof genomic DNA, and H2O up to 10 �l. The numberof cycles and temperature of annealing used for eachpair of primers are given in Table 1. The yield andspecificity of PCR products were evaluated after elec-trophoresis in 2% agarose gel (Gibco) stained with ethid-ium bromide. Then the PCR products were purified withGenElute PCR DNA Purification Kit (Sigma–AldrichCorporation, St. Louis, MO, USA) and sequenced in anABI 377 sequencer (Applied Biosystems, Foster City,CA, USA). Sequences were analysed using Sequencher(Gene Codes Corporation, Ann Arbor, MI, USA) soft-ware.

2.3. Analysis of GHR genotypes

Primers GHR3 (labelled at 5′ end with fluoresceindye C5) and GHR4 (Table 1) were used for the anal-ysis of the TG-repeat (microsatellite) polymorphism.The PCR products were analysed after 5 min denaturingin a 50% formamide solution containing blue dextran.The fluorescent PCR products were separated in 6%denaturing polyacrylamide gels, using an ALFexpressDNA Sequencer (Amersham Biosciences Corporation,Piscataway, NJ, USA). In each lane 1 �l of PCR prod-ucts were resolved together with a size marker. Afterautomated allele calling using the Allele Links 1.01software (Amersham Biosciences), individual genotypeswere checked by manual inspection before exporting thegenotypes to Excel.

2.4. Analysis of milk composition

The goats were milked twice a day. Milk sampleswere taken from each goat once a month during thewhole lactation period. The milk yield and content ofthe major milk components – fat, protein, and lactose

and also somatic cell count (SCC) – were collected. Thedaily milk yield was determined and the fat, protein andlactose content in milk samples were estimated in freshmilk using Milko Scan 104A/B (FOSS A/S, Hillerød,

A. Maj et al. / Small Ruminant Research 67 (2007) 279–284 281

Table 1Sequences of primers and PCR conditions used in this study to amplify different fragments of the caprine GHR gene

Pairs of primers Primer sequences (5′–3′) Position Tann (◦C) Cycles Size of PCR product (bp)

1 GHR1 F: TGCGTGCACAGCAGCTCAACC <1 65 33 743GHR2 R: GGAGGCTCACAAGGCTCAT 692–710

2 GHR3 F: CTGGCGTATGGTCTTTGTCA 625–644 58 34 313GHR4 R: TGGTCTTGCTGCTTTCCTAT 918–937

3G

DFat

2

ptfugstwS

aktFtwwwctsiactd

omMwT

GHR5 F: GTGATTGGGAGGGAGGAAGAGAGHR6 R: CAAGGAGGGAGGGAGGAATAAA

enmark). Somatic cells were counted by means of aossomatic apparatus (FOSS A/S). Somatic cell countnd lactose concentration in milk were used as indica-ors of the health status of the udder.

.5. Statistical calculations

Altogether 2547 records about daily milk yield, fat,rotein for 157 goats with 25 different TG-repeat geno-ypes and 1017 records about lactose content, and SCCor 52 goats with 20 different TG-repeat genotypes weresed in statistical analysis. For calculations only GHRenotypes carried by more than three animals were con-idered. Eighty-eight goats were of the PWI and 69 ofhe PFI breed. The SCC values (expressed in thousands)ere transformed to the natural logarithm scale (ln ofCC).

Milk traits were investigated between 1999 and 2004nd each year constituted a different class of the year ofidding. Three seasons of kidding were established, withhe first class covering kidding from November throughebruary next year, the second one—in March, and the

hird class in the rest of a year. Three classes of parityere distinguished, with the third class comprising goatsith more then two lactations (parity 3–6). The animalsere also grouped according to litter size, with the 1st

lass comprising goats with a single kid and the 2nd withwins or triplets. The combined effect of herd, year andeason of kidding was evaluated by dividing the animalsnto 27 classes for milk yield, fat and protein contentnd 15 classes for lactose content and SCC. Ninety-ninelasses of test day for milk yield, fat and protein con-ent and 55 classes for lactose content and SCC wereistinguished.

In order to determine the impact of the polymorphismf the GHR gene on the investigated traits the REML

ethod was used with repeatability, multi-traits Animalodel based on test day information. The DMU programas used for computation (Madsen and Jansen, 2000).he basic effects in the model were chosen according

853–874 68 33 456>1269

to earlier studies on whole Polish active goat popula-tion (Bagnicka and Łukaszewicz, 1999). The followingmodel was used:

yijklmnop = μ + ai + pi + hsj + GMk + Pl + HYSm

+ LSn + TDo + (Σbp DIMp)ijklmnop

+ β1(x2 − x)ijklmnop + eijklmnop

where yijklmnop is the observed mean value of a trait, μ theoverall mean, ai the random effect of animal, pi the ran-dom permanent environment of animal, hsj the randomeffect of herd-sire, GMk the fixed effect of genotype,Pl the fixed effect of parity, HYSm the fixed effect ofherd-year-season of kidding, LSn the fixed effect of littersize, TDo the fixed effect of test day, (�bpDIMp)ijklmnopthe regressions on days between kidding and milking(DIM), standardized and converted to Legendre polyno-mials (LPs) the up to the fifth power of LPs (p = 1–5),β1(x2 − x)ijklmnop the regression on milk yield for fat,protein and lactose content and log (SCC), and eijklmnopis the random error.

Legendre polynomials are commonly used for test-day models (Kettungen et al., 2000). The effect of breedwas not considered in the model because its impact wasnot significant on any of the investigated traits.

The differences of the observed and expected fre-quencies of genotypes within breeds were tested usingthe Fisher’s exact test (Raymond and Rousset, 1995a,b)to confirm (or deny) that analysed populations were inHardy–Weinberg equilibrium. Moreover, observed fre-quencies of alleles and genotypes between breeds wereestimated using Fisher’s test with the null hypothesisthat allelic and genotypic distribution is identical acrosspopulations. Calculation were performed using GenepopVersion 3.4 (Raymond and Rousset, 1995a,b).

Heterozygosity (H) was calculated according to a for-mula described by Nei (1978); polymorphism informa-tion content (PIC) was calculated according to a formuladescribed by Botstein et al. (1980).

ant Research 67 (2007) 279–284

Table 3Number of records, means and standard deviations of investigated traits

Trait N Mean S.D.

Daily milk yield (kg) 2547 2.38 1.04Fat (%) 2547 3.35 1.10Protein (%) 2547 3.26 0.81

282 A. Maj et al. / Small Rumin

3. Results

Basing on the sequences of the bovine and ovine GHRgenes available in the GenBank database, we designedthree pairs of PCR primers enabling amplification ofthe 5′-region of the caprine GHR gene. Using theseprimers we amplified and then sequenced three overlap-ping fragments of the 5′-region. Altogether, these frag-ments were combined into a 1269-bp sequence whichwas deposited in the GenBank database under the acces-sion no. AY358031.

A (TG)14-repeat was found within the caprine GHRgene 5′-noncoding region at position 828–855, starting84 bp upstream from the putative transcription initiationsite. This region corresponded roughly to promoter P1preceding exon 1A in the bovine GHR gene. Moreover,a 317-bp long interspersed repetitive element (LINE-1)was identified at location 482 bp upstream from the tran-scription initiation site.

Further analysis of the GHR gene 5′-region in 235goats from two breeds (PWI and PFI) showed that thelength of the TG-repeat differed between individuals. Atotal of 10 alleles of the GHR gene were detected in thetwo breeds, the length of the amplified fragment contain-ing the variable TG-repeat ranging from 305 to 346 bp(Table 2). The frequency of the alleles varied between0.01 and 0.48, allele 2 (313 bp) being the most frequentin both breeds. Among the 10 alleles identified, eightoccurred in homozygous systems. The frequency of the

homozygous genotypes was 0.25 for the PWI and 0.28for the PFI breed. The mean heterozygosity (H) and poly-morphic content (PIC) coefficients at this locus weresimilar for both breeds (Table 2).

Table 2Allele frequencies in GHR gene and mean heterozygosity (H) and polymorphhypothesis that the allelic distribution is identical across populations

Allele Length ofallele (bp)

Polish whiteimproved (n = 115)a

1 305 0.072 313 0.483 320 0.034 327 0.115 331 0.026 333 0.087 336 0.108 338 0.049 341 0.0110 346 0.06Fisher’s exact test P < 0.01H 0.74PIC 0.71

a n: Number of animals tested.

Lactose (%) 1017 4.39 0.32SCC (thousand) 1017 1756 2385SCC (ln) 1017 6.76 1.10

No significant difference between the expected andobserved frequencies of genotypes was noted (Fig. 1).Differences in the genotype and allele frequencies weresignificant (P < 0.01) between the breeds. CodominantMendelian inheritance of TG-repeat alleles was observedin two goat families (Fig. 2).

Data about milk yield, fat and protein content derivedfrom 157 does and about lactose content and SCC from52 does were used for association studies. The numberof observations, mean values, and standard deviations ofthe investigated traits are shown in Table 3. The aver-age daily milk yield during lactation and the contentsof milk components were rather high. A high variationwas observed for milk yield. For fat content the varia-tion was medium while for protein and lactose content itwas low. The average somatic cell count and its variationwere very high, indicating probably that some goats hadproblems with subclinical mastitis.

No associations were found between TG-repeat

length polymorphism and dairy traits – milk yield andcontent of the major milk components – fat, protein, andlactose (not shown). Also, no effect was shown of theGHR genotype on the somatic cell count (SCC).

ic content (PIC) in goats of two breeds, Fisher’s exact test of the null

Polish fawnimproved (n = 120)a

Combined for twobreeds (n = 235)a

0.08 0.080.48 0.480.07 0.050.01 0.060.06 0.040.03 0.060.05 0.080.07 0.050.08 0.040.07 0.07

0.74 0.740.72 0.72

A. Maj et al. / Small Ruminant Research 67 (2007) 279–284 283

F e) genots sa reprer 20-bp h

4

Gm(attdo(

ig. 1. Observed and expected frequencies of TG-repeat (microsatellitymbols were used for the GHR genotypes; the numbers under abscisepresent heterozygote with alleles 305 and 346 bp; “3/3” represents 3

. Discussion

A TG-repeat occurs in homologous positions of theHR gene 5′-noncoding region in most studied mam-alian species—human (Pekhletsky et al., 1997), mouse

Menon et al., 1995), sheep (O’Mahoney et al., 1994),nd cattle (Lucy et al., 1998). The repeat was foundo be polymorphic in cattle (Lucy et al., 1998) and in

he European bison (Bison bonasus; our unpublishedata). The GT-repeat polymorphism in the P1 promoterf bovine GHR gene was first described by Lucy et al.1998). Five alleles were found with variable TG-repeat

Fig. 2. Mendelian inheritance of GHR gene TG-repeat (micros

ypes within the 5′-noncoding region of the goat GHR gene. Numericalsent symbols of alleles of the lengths showed in Table 2. Thus, “1/0”omozygote, etc.

number. Later on, Hale et al. (2000) reported an asso-ciation between the microsatellite marker and growthrates in Angus steers (weaning weight and carcassweight).

The present study, performed on 235 goats fromtwo breeds (Polish White Improved and Polish FawnImproved) for the first time showed the presence of avariable TG-repeat within the caprine GHR gene 5′-

noncoding regions. The position of the microsatellitewas roughly the same as in the bovine GHR gene—theP1 promoter preceding exon 1A. Ten alleles were foundwith the length of the amplified fragment ranging from

atellite) alleles and genotypes in two families of goats.

ant Re

284 A. Maj et al. / Small Rumin

305 to 346 bp (approximately 10–30 TG-repeats, respec-tively).

Although microsatellites are most often considered tobe neutral markers, recent evidence suggests that poly-TG elements might have functional significance. Theyoften adopt a left-handed double helical conformationof DNA called Z-DNA that may be important in muta-genesis, recombination, and control of gene expression(Majewski and Ott, 2000). The length of the GT-repeatshas been shown to influence gene transcription rates ofseveral human genes, including EGF receptor, matrixmetalloproteinase-9, and type I collagen �2; in all thecases gene transcription rates were positively associatedwith the length of the GT-repeat (Hadjiyannakis et al.,2001).

In this study, a novel short tandem repeat (STR) poly-morphism was found within the 5′-noncoding region ofthe goat GHR gene. Also, for the first time the associ-ations were studied between GHR gene polymorphismand goat production traits. However, our data showedthat the GT-repeat polymorphism in the GHR gene hasnot effect on the goat dairy traits under study. There-fore, the TG-repeat polymorphism in P1 promoter of thegoat GHR gene does not seem a useful marker for milkproduction traits.

Acknowledgements

This study was funded by the Ministry of Educationand Science, Poland, grants PBZ-KBN-036/P06/12 and3 P06D 02525 and the IGHZ project S.I.-1.2.

References

Aggrey, S.E., Yao, J., Sabour, M.P., Lin, C.Y., Zadworny, D., Hayes,J.F., Kunlein, U., 1999. Markers within the regulatory region ofthe growth hormone receptor gene and their association with milk-related traits in Holsteins. J. Hered. 90, 148–151.

Bagnicka, E., Łukaszewicz, M., 1999. Genetic and environmental vari-ation of dairy traits in Polish goats. Anim. Sci. Pap. Rep. 17, 59–65.

Blott, S., Kim, J.J., Moisio, S., Schmidt-Kuntzel, A., Cornet, A., Berzi,P., Cambisano, N., Ford, C., Grisart, B., Johnson, D., Karim, L.,Simon, P., Snell, R., Spelman, R., Wong, J., Vilkki, J., Georges,M., Farnir, F., Coppieters, W., 2003. Molecular dissection of aquantitative trait locus. A phenylalanine-to-tyrosine substitution inthe transmembrane domain of the bovine growth hormone receptoris associated with a major effect on milk yield and composition.Genetics 163, 253–266.

Botstein, D., White, R.L., Skalnick, M.H., Davies, R.W., 1980. Con-

struction of a genetic linkage map in man using restriction fragmentlength polymorphism. Am. J. Hum. Genet. 32, 314–331.

Burton, J.L., McBride, B.W., Block, E., Glimm, D.R., Kenelly, J.J.,1994. A review of bovine growth hormone. Can. J. Anim. Sci. 74,167–201.

search 67 (2007) 279–284

Etherton, T.D., Bauman, D.E., 1998. Biology of somatotropin ingrowth and lactation of domestic animals. Physiol. Rev. 78,745–761.

Falaki, M., Gengler, N., Sneyers, M., Prandi, A., Massart, S.,Formigoni, A., Burny, A., Portetelle, D., Renaville, R., 1996.Relationships of polymorphisms for growth hormone and growthhormone receptor genes with milk production traits for ItalianHolstein–Friesian bulls. J. Dairy Sci. 79, 1446–1453.

Hadjiyannakis, S., Zheng, H., Hendy, G.N., Goodyer, C.G., 2001. GT-repeat polymorphism in the 5′-flanking region of the human growthhormone receptor gene. Mol. Cell. Probes 15, 239–242.

Hale, C.S., Herring, W.O., Shibuya, H., Lucy, M.C., Lubahn, D.B.,Keisler, D.H., Johnsson, G.S., 2000. Decreased growth in Angussteers with a short TG-microsatellite allele in the P1 promoter ofthe growth hormone receptor gene. J. Anim. Sci. 78, 2099–2104.

Jarrige, R. (Ed.), 1988. Alimentation Des Bovins, Ovins and Caprins.INRA, Paris, p. 465.

Jiang, H., Lucy, M.C., 2001. Variants of the 5′-untranslated region ofthe bovine growth hormone receptor mRNA: isolation, expressionand effects on translational efficiency. Gene 265, 45–53.

Jiang, H., Okamura, C.S., Lucy, M.C., 1999. Isolation and characteri-zation of a novel promoter for the bovine growth hormone receptorgene. J. Biol. Chem. 274, 7893–7900.

Kanai, N., Fujii, T., Saito, K., Yokoyama, T., 1994. Rapid and sim-ple method for preparation of genomic DNA from easily obtainedclotted blood. J. Clin. Pathol. 47, 1043–1044.

Kettungen, A., Mantysaari, E.A., Poso, J., 2000. Estimation of geneticparameters for daily milk yield of primiparous Ayrshire cows byrandom regression test-day models. Livest. Prod. Sci. 66, 251–261.

Lucy, M.C., Johnsson, G.S., Shibuya, H., Boyd, C.K., Herring,W.O., Werin, M., 1998. Rapid communication: polymorphic (GT)n

microsatellite in the bovine somatotropine receptor gene promoter.J. Anim. Sci. 76, 2209–2210.

Madsen, P., Jansen, J., 2000. A user’s guide to DMU. A package foranalysing multivariate mixed models. Version 6. Release 4.

Majewski, J., Ott, J., 2000. GT-repeats are associated with recom-bination on human chromosome 22. Genome Res. 10, 1108–1114.

Menon, R.K., Stephan, D.A., Manbir, S., Morris, S.M., Zou, L., 1995.Cloning the promoter-regulatory region on the murine growth hor-mone receptor. J. Biol. Chem. 270, 8851–8859.

Nei, M., 1978. Estimation of average heterozygosity and genetic dis-tance from a small number of individuals. Genetics 89, 583–590.

O’Mahoney, J.V., Brandon, M.R., Adams, T.E., 1994. Identification ofa liver-specific promoter for the ovine growth hormone receptor.Mol. Cell. Endocrinol. 101, 129–139.

Parmentier, I., Portetelle, D., Gengler, N., Prandi, A., Bertozzi, C.,Vleurick, L., Gilson, R., Renaville, R., 1999. Candidate gene mark-ers associated with somatotropic axis and milk selection. Domest.Anim. Endocrinol. 17, 139–148.

Pekhletsky, R.I., Chernov, B.K., Rubtsov, P.M., 1997. Variants of the 5′-untranslated sequence of human growth hormone receptor mRNA.Mol. Cell. Endocrinol. 90, 103–109.

Raymond, M., Rousset, F., 1995a. An exact test for population differ-entiation. Evolution 49, 1280–1283.

Raymond, M., Rousset, F., 1995b. Genepop (version 1.2): population

genetics software for exact tests and ecumenicism. J. Hered. 86,248–249.

Zhu, T., Goh, E.L.K., Graichen, R., Ling, L., Lobie, P.E., 2001. Signaltransduction via the growth hormone receptor. Cell. Signal. 13,599–616.