A mammary gland EST showing linkage disequilibrium to a milk
production QTL on bovine Chromosome 14
Institut fu¨r Tierzucht und Tierhaltung der Christian-Albrechts-Universita¨t, D-24098 Kiel, Germany
Forschungsinstitut fu¨r die Biologie landwirtschaftlicher Nutztiere, Forschungsbereich Molekularbiologie, D-18196 Dummerstorf, Germany
Received 20 October 2000 / Accepted: 11 April 2001
Abstract. As part of a genome scan, ESTs derived from mam-
mary gland tissue of a lactating cow were used as candidate genes
for quantitative trait loci (QTL), affecting milk production traits.
Resource families were genotyped with 247 microsatellite markers
and 4 polymorphic ESTs. It was shown by linkage analysis that
one of these ESTs, KIEL_E8, mapped to the centromeric region of
bovine Chromosome (Chr) 14. Regression analysis revealed the
presence of a QTL, with significant effect on milk production, in
this chromosome region, and analysis of variance showed no sig-
nificant interaction of marker genotype and family. The estimated
significant differences between homozygous marker genotypes
were 140 kg milk, −5.02 kg fat yield, and 2.58 kg protein yield for
the first 100 days of lactation. Thus, there was strong evidence for
a complete or nearly complete linkage disequilibrium between
KIEL_E8 and the QTL. To identify the biological function of
KIEL_E8, we extended the sequence for 869 bp by 5Ј-RACE. A
560-bp fragment of this shows a 90.9% similarity to a gene en-
coding a cysteine- and histidine-rich cytoplasmic protein in mouse.
Although such a protein may have a regulatory function for lac-
tation and a linkage disequilibrium between the EST marker and
the QTL has been observed, it remains to be elucidated whether
they are identical or not. Nevertheless, KIEL_E8 will be an effi-
cient marker to perform marker-assisted selection in the Holstein-
Several QTL associated with some milk production traits have
been identified by genome screening with microsatellite markers
(Georges et al. 1995), but it is only the milk protein genes that have
been cloned and thoroughly characterized (Threadgill and Wom-
ack 1990). Thus, dissecting the molecular basis of a QTL for milk
production traits remains one of the most challenging tasks in
A QTL with a major effect on milk yield and composition was
reported in the centromeric region of Chr 14 (Coppieters et al.
1998; Heyen et al. 1999). Because of its large effect, this QTL
bears particular significance for the application of fine mapping
techniques and subsequent positional cloning approach.
Although the positional cloning strategy has been successful in
isolating a number of human disease genes (Collins 1995) and
recently also in pigs (Milan et al. 2000), it has often been supple-
mented successfully with the mapping of candidate genes for the
trait of interest, an approach often referred to as a positional can-
didate gene approach (Collins 1995). In domestic animals, where
gene maps are relatively poorly developed, this approach is often
combined with extrapolating the dense gene maps of human and
mouse by using the knowledge of evolutionarily conserved syn-
teny groups. An example from pigs is the genetic mapping of the
white coat color phenotype to Chr 8 and subsequent identification
of the KIT gene as the underlying mutated gene (Johansson Moller
et al. 1996). For traits where substantial physiological differences
exist between species, it may, however, be relevant to develop
transcript maps for the species of interest.
Up to now, few EST-projects have been reported in livestock
species, and only two used mammary gland tissue. Le Provost et
al. (1996) reported 140 genes transcribed in goat udder, and we
recently sequenced and mapped 16 ESTs derived from mammary
gland tissue (Karall-Albrecht et al. 2000). Because of their differ-
ential expression in a lactating mammary gland, we have the hy-
pothesis that these ESTs are candidate genes for milk production
traits. One of these markers was genetically mapped with the In-
ternational Bovine Reference Panel (IBRP) families to the centro-
meric region of Chr 14, and it may, therefore, be a positional
candidate gene for the reported milk production QTL.
In the present study, we have tested 12 bovine mammary gland
ESTs as candidate genes by integrating the polymorphic markers
in a genome scan with a granddaughter design to map milk pro-
duction QTL. Results of the genome scan with relevance to the
EST-markers, significant estimated effects of one marker, and its
further characterization are presented.
Materials and methods
Resource population and recorded traits.
The resource families were
established within the bovine genome mapping project of the German
Cattle Breeders Association (ADR, Bonn, Germany). German AI-
organizations contributed 1393 semen samples from 22 paternal half-sib
families, 18 German Holsteins, three German Simmentals, and 1 Brown
Swiss family. The number of sons per sire ranged from 19 to 127 and was
on average 53.7. Additionally, semen from three grandsires was available.
Data were obtained from the United Datasystems for Animal Production
(VIT, Verden, Germany) and the Bavarian Institute of Animal Breeding
(Grub, Germany) for the following milk production traits: milk, fat, and
protein yield, as well as fat and protein percentage. Breeding values for the
first 100 days of lactation, based on at least 25 daughter records, were used
for statistical analysis.
Marker genotyping and linkage analysis.
Microsatellite typing of 247
markers was performed as previously described by Thomsen et al. (2000).
The microsatellite marker ILSTS039 (http://spinal.tag.csiro.au/cgi-bin/
cgdlose?ILSTS39) and twelve polymorphic ESTs, expressed in mammary
gland tissue (Karall et al. 1997a, 1997b; Karall-Albrecht et al. 2000) were
also used for genotyping. PCR-conditions, primer sequences, and condi-
tions for F-SSCP-analysis were described in the references mentioned
above. All genotypic data were transformed to the ADRDB database and
checked for typing errors by using software that analyzed the scored geno-
types for Mendelian segregation (Reinsch 1999).
Correspondence to: Christian Looft; E-mail: firstname.lastname@example.org
Mammalian Genome 12, 646–650 (2001).
© Springer-Verlag New York Inc. 2001