1Federal University of Sa ˜o Carlos, Sa ˜o Carlos,2Animal Science Institute of Sa ˜o Paulo, Nova Odessa,
3Embrapa Southeast Cattle Research Center, Sa ˜o Carlos and4Animal Science Department – ESALQ/
USP, Piracicaba, SP, Brazil
Candidate genes for growth traits in beef cattle crosses
Bos taurus · Bos indicus
By D. D. Tambasco1, C. C. P. Paz2, M. Tambasco-Studart1, A. P. Pereira1,
M. M. Alencar3, A. R. Freitas3, L. L. Coutinho4, I. U. Packer4and L. C. A. Regitano3
Candidate gene strategy has been proposed to direct the search for quantitative trait loci, and
polymorphisms at these loci have been associated with several production traits in bovine. A total of
211 animals resulting from crosses between Nelore females with sires from either Aberdeen Angus,
Canchim and Simmental breeds were analyzed for restriction fragment length polymorphisms
(RFLPs) in the j-casein, b-lactoglobulin (LGB) and growth hormone GH genes. Average daily gain
from birth to weaning (GBW) and from weaning to yearling (GWY) were analyzed by a statistical
model comprising RFLP genotypes and their interaction effects. Significant effects (p < 0.05) of
growth hormone (GH) and of the interaction b-lactoglobulin · GH were verified for the GWY.
Animals with LL genotypes for GH had higher GBW than LV animals but lower GWY (p < 0.05).
Kandidatengene fu ¨r Wachstum in Fleischrinderkreuzungen Bos taurus · Bos indicus
Fu ¨r die direkte Suche nach QTLs ist der Kandidatengen-Ansatz eine Mo ¨glichkeit und Polymorphis-
men in diesen Genen zeigen Assoziationen zu Produktionsmerkmalen beim Rind auf. Insgesamt
wurden 211 Tiere aus Kreuzungen zwischen Nelore-Ku ¨hen und Aberdeen Angus-, Canchim- oder
Simmental-Bullen fu ¨r die Analyse von RFLPs im j-Kasein, b-Laktoglobulin (LGB) und Wachstums-
hormon (GH) verwendet. Die durchschnittliche ta ¨gliche Zunahme von der Geburt bis zum Absetzen
(GBW) und vom Absetzen bis zum Alter von einem Jahr (GWY) wurden durch ein statistisches
Modell unter Beru ¨cksichtigung der RFLP-Genotypen und deren Interaktionen analysiert. Signifikante
Effekte (p < 0,05) von GH und die Interaktion b-LGB · GH wurden fu ¨r GWY verifiziert. Tiere mit
dem Genotyp GH · LL hatten ho ¨here GBWs als LV-Tiere, jedoch niedrigere GWY (p < 0,05).
Animal improvement has been achieved by selection based on either phenotype or
predicted additive genetic merit of the superior animals for production traits. Molecular
biology techniques allow the identification of genetic variation at specific loci and the
association between variation at quantitative trait loci (QTL) and production traits. The
final goal is to use marker assisted selection to improve the genetic gain achieved by
selection as a result of higher accuracy on the estimation of an animal’s genetic value.
Candidate gene strategy has been proposed to direct the search for QTL, assuming that the
genetic variation at genes affecting the physiological pathways related to a phenotype
would be more likely to affect the quantitative variation in that phenotype than genes or
chromosome regions chosen by chance. Polymorphisms at the growth hormone gene have
been associated with several production traits in cattle such as milk production and quality
(Lagziel et al. 1996), growth (Rocha et al. 1992, Unanian et al. 2000) and carcass
composition and quality (Schlee et al. 1994a, b; Taylor et al. 1998). Genes coding for
J. Anim. Breed. Genet. 120 (2003), 51–56
? 2003 Blackwell Verlag, Berlin
Ms. received: 02.05.2002
Ms. accepted: 08.11.2002
U.S. Copyright Clearance Center Code Statement: 0931–2668/2003/2001–0051 $15.00/0 www.blackwell.de/synergy
milk proteins have been associated with milk quality and yield (Ron et al. 1994) as well
as growth traits (Lin et al. 1992, Moody et al. 1996). The objective of the present work
was to investigate the effects of growth hormone (GH) – AluI, j-casein-HinfI (CSN3)
b-lactoglobulin (LGB) – HaeIII polymorphism on growth traits of three beef
cattle crosses: Aberdeen Angus · Nelore (STAI), Canchim · Nelore (SRCI) and
Simmental · Nelore (STSI).
Materials and methods
The animals resulted from the following crosses: Nelore females sired by five Aberdeen
Angus (n ¼ 72), five Canchim (n ¼ 74) and four Simmental bulls (n ¼ 65), totaling 211
progenies born in 1998 and 1999. The animals were maintained under intensive
management on Brachiaria brizantha and Panicum maximum pastures for dry and rainy
seasons, respectively. The birth weight means for 1998 and 1999 were 32.9 and 34.4 kg for
STAI, 32.1 and 32.8 kg for SRCI, 34.1 and 33.2 kg for STSI, respectively. The age at
weaning was approximately 240 days. Nelore females used in the crosses were a
representative sample of the Brazilian herd comprising animals from different farms. The
cows used in both reproduction periods were the same, with exception to some
replacements. Their ages at calving ranged from 3.4 to 17.6 years, with a mean of 6.6 years.
The restriction fragment length polymorphisms j-casein – HinfI (CSN3), (LGB) –
HaeIII and (GH) – AluI, located on the chromosomes 6, 11 and 19, respectively, were
determined by digestion of polymerase chain reaction (PCR) products. DNA samples were
obtained from white blood cells using a standard salting out technique. PCR reactions
consisted of 200 ng genomic DNA in a standard PCR buffer with 1.5 mM MgCl2, 200 nM
of each dNTP, 0.4 lM of each primer and 0.5 units Taq DNA polymerase in 25 ll
reactions. Primers for CSN3, LGB and GH were those described by Medrano &
Cordova (1990), Ron et al. (1994) and Schlee et al. (1994a), respectively. PCR products
of CSN3, LGB and GH were digested with HinfI, HaeIII and AluI, respectively and
separated the A and B alleles for the first two loci and the L and V alleles for GH.
Average daily gain from birth to weaning (GBW) and from weaning to yearling age (GWY)
were analyzed by SAS program (1999) using the GLM (General Linear Models)
procedure and at first the following statistical model:
yijklmno¼ l þ GGiþ SðGGÞjþ CGkþ GHlþ LGBmþ CSN3nþ GH ? LGBlm
þ GH ? CSN3lnþ LGB ? CSN3mnþ eijklmno
where, yijklmno¼ trait (GBW and GWY) evaluated on the ijklmnoth animal, l ¼ overall
mean for each trait, GGi, S(GG)j, CGk, GHl, LGBm, CSN3n, GH · LGBlm,
GH · CSN3ln, LGB · CSN3mn¼ fixed effects associated with the ith genetic group, the
jth sire nested within genetic group, the kth contemporary group, the lth GH genotype (LL,
respective genotypes. eijklmno¼ random error associated with ijklmnoth observation.
This model was used as a preliminary analysis and the non-significant effects for CSN3
and corresponding interactions were discarded for the final analysis.
Results and discussion
The three loci were polymorphic in all genetic groups. The genotypes AA/AB/BB for the
milk proteins polymorphisms and LL/LV for the GH locus were observed in all genetic
D. D. Tambasco et al.
groups. Despite other results suggesting effects of CSN3 genotype on growth traits
(Moody et al. 1996; Lin et al. 1992), the present data did not show any influence of CSN3
genotypes on the traits studied. Therefore, this main effect and its respective interactions
were excluded from the statistical model. Significant differences (p < 0.01) on GBW
trait were observed for GG, CG, sire (GG) and GH effects: for the GG effects these
differences were Aberdeen Angus · Nelore vs. Canchim · Nelore and Canchim · Nelore
vs. Simmental · Nelore (data not shown). For the second period of growth (GWY) CG,
GH and the interaction LGB · GH contributed significantly to the variation (Table 1).
LGB alone did not influence GBW and GWY but there was a significant interaction
(p < 0.05) of this locus with GH for GWY and an approximate significance for GBW
(Table 1). Animals with LL genotypes for GH had higher GBW (p < 0.05) than LV
animals but lower GWY (Table 2). As all Nelore females were homozygotes LL, no VV
genotype was observed in the progenies of these crosses, so it was not possible to
determine the type of gene action. Previous studies have found influence of this
polymorphism on growth traits. Genotype LV was associated with a higher carcass gain
than LL and VV in Simmental sires (Schlee et al. 1994a, b) whereas homozygotes VV had
better carcass classification scores (Schlee et al. 1994a). Working on Piemontese ·
Chianina crosses, Pilla et al. (1994) found the VV genotype to be associated with the
highest chest girth and the lowest chest length. Sartori and Di Stasio (2000) suggested
indirect selection for the V allele, associated with preselection after a performance test in
Piemontese calves. However, in a later work (Di Stasio et al. 2002), the authors did not
provide evidence of associations of GH polymorphisms with meat production traits in
Table 1. ANOVA summary for GBW and GWY
Source DFGBW Mean squareGWY Mean square
GH · LGB
** p £ 0.01; * p £ 0.05; GG, CG ¼ genetic group, contemporary group, respectively.
GBW and GWY means and standard error (SE) according to GH genotype and
GH · LGB combinations
GH LGB GBW (kg) Mean ± SEGWY (kg) Mean ± SE
0.835 ± 0.014
0.784 ± 0.020
0.832 ± 0.017
0.809 ± 0.017
0.862 ± 0.014
0.764 ± 0.043
0.809 ± 0.024
0.777 ± 0.025
0.651 ± 0.016
0.735 ± 0.024
0.631 ± 0.035
0.671 ± 0.019
0.649 ± 0.017
0.792 ± 0.051
0.678 ± 0.029
0.735 ± 0.030
GBW, GWY ¼ average daily gain from birth to weaning and from weaning to yearling, respectively.
Candidate genes in beef cattle crosses
In the present work, differences for GWY between LL and LV genotypes were
significant only when animals were homozygotes AA or BB for LGB (Fig. 1). Figure 2
shows the interactions between GH and LGB genotypes for GBW. For this period the
interaction was not significant, although statistical significance is barely missed. Means of
heterozygotes for GH with respect to LGB genotypes were 0.79, 0.68 and 0.73 kg for AA,
BB and AB genotypes, respectively. The only significant difference (p < 0.05) was
observed between the first two means. Among LL individuals there were no differences
with respect to LGB genotype.
The same analysis was conducted to verify the effects of GH and LGB genotypes on
the weight gain from birth to yearling, as a sum of the two growth periods. The results
demonstrated the non-significance of these genotypes, as well of their interaction. This
fact could be related to the opposite effects of GH genotypes on the GBW and GWY
Interactions between loci (epistasis) have not been considered for quantitative traits
analysis for practical reasons, including mathematical modeling and computational power.
The lack of practical evidences of its existence has long justified this simplification. This
scenario is being changed by recent findings. Piccoli et al. (2002) observed epistatic effects
on pre-weaning gain of Hereford · Nelore calves. Similar effects were verified on the post-
weaning weight gain of Nelore calves (Cardoso et al. 2002). With the advances in
molecular markers studies, individual contributions of QTLs can be isolated, and
interactions may be discovered. Recently, Casas et al. (2000) identified an interaction
between myostatin and two QTLs, on chromosome 5 and for fat depth on chromosome 14
for Warner-Bratzler shear force. Cheverud (2002), searching for QTLs affecting growth in
inbred mice strains, found many instances of epistatic interaction among loci affecting this
trait. Using mouse models to dissect the genetics of obesity, Brockmann and Bevova
(2002) observed that interactions between QTLs contributed considerably to the
phenotypic variance of growth and obesity.
Fig. 1. Interaction of LGB and GH genotypes for GWY. (Axis X ¼ LGB genotypes, Axis Y ¼
D. D. Tambasco et al.
In most QTL studies, statistical analyses have routinely considered each locus separately.
Thus, they do not account for interaction between different loci. The present study
indicates that interactions between marker loci should be more carefully investigated as
such interactions may account for differences in genotype responses across populations
and genetic backgrounds. The growth hormone LV genotype was positively related to
gain at later ages (GWY), which is advantageous for beef cattle improvement, however,
the influence of the LGB genotype should be taken into consideration.
This work had financial support from FAPESP, Embrapa and CNPq. LL Coutinho is a recipient of a
productivity scholarship from CNPq.
Brockmann, G. A.; Bevova, M. R., 2002: Using mouse models to dissect the genetics of obesity.
Trends Genet. 18: 367–376.
Cardoso, V.; Queiroz, S. A.; Brito, F. V.; Fries, L. A., 2002: Evidence of heterotic and epistatic
effects on postweaning weight gain of Nelore calves. In: 7th World Congress on Genetics Applied
to Livestock production.
Casas, E.; Shackelford, S. D.; Keele, J. W.; Stone, R. T.; Kappes, S. M.; Koohmaraie, M., 2000:
Quantitative trait loci affecting growth and carcass composition of cattle segregating alternate
forms of myostatin. J. Anim. Sci. 78: 560–569.
Cheverud, J. M., 2002: Quantitative trait loci affecting postnatal growth in mice. In: 7th World
Congress on Genetics Applied to Livestock production.
DiStasio, L.; Sartore, S.; Albera, A., 2002: Lack of association of GH1 and POU1F1 gene variants
with meat production traits in Piemontese cattle. Anim. Genet. 33: 61–64.
Fig. 2. Interaction of LGB and GH genotypes for GBW. (Axis X ¼ LGB genotypes, Axis Y ¼
Candidate genes in beef cattle crosses
Lagziel, A.; Lipkin, E.; Soller, M., 1996: Association between SSCP haplotypes at the bovine Download full-text
growth hormone gene and milk protein percentage. Genetics 142: 945–951.
Lin, C. Y.; Sabour, M. P.; Lee, A. J., 1992: Direct typing of milk proteins as an aid for genetic
improvement of dairy bulls and cows: a review. Anim. Breed. Abst. 60: 1–10.
Medrano, J. F.; Aguilar-Cordova, E., 1990: Genotyping of bovine kappa-casein loci following
DNA sequence amplification. Bio/Technology 8: 144–146.
Moody, D. E.; Pomp, D.; Newman, S.; McNeil, M. D., 1996: Characterization of DNA
polymorphisms in three populations of Hereford cattle and their associations with growth and
maternal EPD in line 1 Herefords. J. Anim. Sci. 74: 1784–1793.
Piccoli, M. L.; Roso, V. M.; Brito, F. V.; Severo, J. L. P.; Schenkel, F. S.; Fries, L. A.., 2002.
Additive, complementarity, (additive*additive), dominance, and epistatic effects on preweaning
weight gain of Hereford · Nelore calves. In: 7th World Congress on Genetics Applied to
Pilla, A. M.; Napolitano, F.; Moioli, B. M.; Puppo, S.; Pilla, F.; Carretta, A.., 1994: Association
between restriction fragment length polymorphisms and quantitative traits in Piemontese · Chi-
anina crossbred. In: Proceedings of the 5th World Congress on Genetics Applied to Livestock
Production 21, pp. 284–287.
Rocha, J. L.; Baker, J. F.; Womack, J. E.; Sanders, J. O.; Taylor, J. F., 1992: Statistical associations
between restriction fragment length polymorphisms and quantitative traits in beef cattle. J. Anim.
Sci. 70: 3360–3370.
Ron, M.; Yoffe, O.; Ezra, E.; Medrano, J. F.; Weller, J. I., 1994: Determination of effects of milk
protein genotyping on production traits of Israeli Holstein. J. Dairy Sci. 77: 1106–1113.
Sartori, S.; Di Stasio, L., 2000: Analisi genetica del locus GH (ormone della crescita) nella razza
bovina Piemontesa. In: Proceedings of the 54th Convegno Societa ` Italiana delle Scienze
Veterinarie LIV. pp. 411–412.
SAS Institue Inc., 1999. SAS/STAT User’s Guide: SAS Institute Inc., Cary, NC, USA.
Schlee, P.; Graml, R.; Rottmann, O.; Pirchner, F., 1994a: Influence of growth – hormone
genotypes on breeding values of Simmental bulls. J. Anim. Breed. Genet. 111: 253–256.
Schlee, P.; Graml, R.; Schallenberger, E.; SChams, D.; Rottmann, O.; Olbrich-Bludau, A.;
Pirchner, F., 1994b: Growth hormone and insulin-like growth factor-I concentrations in bulls of
various growth hormone genotypes. Theor. Appl. Genet. 88: 497–500.
Taylor, J. F.; Coutinho, L. L.; Herring, K. L., Gallagher Jr, D. S.; Brenneman, R. A.; Burney,
N.; Sanders, J. O.; Turner, J. W.; Smith, S. B.; Miller, R. K.; Savell, J. W.; Davis, S. K., 1998:
Candidate Gene Analysis of Gh1 for effects on growth and carcass composition of cattle. Anim.
Genet. 29: 194–201.
Unanian, M. M.; Barreto, C. C.; Freitas, A. R.; Cordeiro, C. M. T.; Josahkian, L. A., 2000:
Associac ¸a ˜o do polimorfismo do gene do hormo ˆnio de crescimento com a caracterı ´stica peso em
bovinos da rac ¸a Nelore. Rev. Bras. Zootec. 29: 1380–1386.
Authors’ addresses: D. D. Tambasco, M. Tambasco-Studart and A. P. Pereira, Federal University
of Sa ˜o Carlos, Box 678, 13565-905, Sa ˜o Carlos, SP, Brazil;
C. C. P. Paz, Animal Science Institute of Sa ˜o Paulo, Box 60, 13400-000, Nova
Odessa, SP, Brazil;
M. M. Alencar, A. R. Freitas and L. C. A. Regitano (for correspondence),
Embrapa Southeast Cattle Research Center, Box 339, 13560-970 Sa ˜o Carlos, SP,
L. L. Coutinho, I. U. Packer, Animal Science Department – ESALQ/USP, Box
09, 13418-900, Piracicaba, SP, Brazil. E-mail: firstname.lastname@example.org
D. D. Tambasco et al.