The Belt mutation in pigs is an allele at the Dominant white
Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, Uppsala Biomedical Centre, Box 597,
S-751 24 Uppsala, Sweden
Molecular Biology Group, Pig Improvement Company, Cambridge CB1 2JN, UK
Deutsche PIG Improvement GmbH, Ratsteich 31, D-24837 Schleswig, Germany
Received: 5 May 1999 / Accepted: 3 August 1999
Abstract. A white belt is a common coat color phenotype in pigs
and is determined by a dominant allele (Be). Here we present the
result of a genome scan performed using a Hampshire (Belt)/
Pietrain (non-Belt) backcross segregating for the white belt trait.
We demonstrate that Belt maps to the centromeric region of pig
Chromosome (Chr) 8 harboring the Dominant white (I/KIT) locus.
Complete cosegregation between Belt and a single nucleotide
polymorphism in the KIT gene was observed. Another potential
candidate gene, the endothelin receptor type A gene (EDNRA),
was excluded as it was assigned to a different region (SSC8q21) by
FISH analysis. We argue that Belt is a regulatory KIT mutation on
the basis of comparative data on mouse KIT mutants and our
previous sequence analysis of the KIT coding sequence from a
Hampshire pig. Quantitative PCR analysis revealed that Belt is not
associated with a KIT duplication, as is the case for the Patch and
Dominant white alleles. Thus, Belt is a fourth allele at the Dom-
inant white locus, and we suggest that it is denoted I
A white belt across the shoulders and front legs is a common coat
color phenotype in the pig and is characteristic of the Hampshire
breed (Fig. 1). It may occur against a solid black (Hampshire,
Essex, and Wessex Saddleback breeds) or red background (Bavar-
ian Landschwein). The inheritance of the white belt was first stud-
ied by Spillman (1907), who assumed it to be owing to the comple-
mentary action of two genes, whereas later studies revealed a
major dominant gene denoted Be (Ollivier and Sellier 1982). How-
ever, the genetic basis of the trait in different breeds, as well as the
interpretation of the variation in width and position of the belt
across the body axis (mainly observed in Wessex Saddleback and
Essex pigs), is still under debate. It has been proposed that white
belts can be obtained within some pied breeds (Meishan, Jinhua)
simply by performing artificial selection in favor of the extension
of the white pattern (Legault 1998).
Two of the major coat color loci in the pig have previously
been investigated at the molecular level, the Dominant white
(I/KIT) locus (Johansson-Moller et al. 1996; Marklund et al. 1998)
and the Extension/MC1R locus (Kijas et al. 1998). Three alleles
have been documented at the former locus: Dominant white (I),
), and the recessive wild type allele (i) (Johansson et al.
1992). Both I and I
are associated with a long-range duplication
that contains the entire coding sequence of KIT (Marklund et al.
1998). Five E/MC1R alleles were found: E
nant black, E
for wild-type color, E
for black spotting, and e for
Mouse is the prime model for coat color genetics in mammals.
Many mice mutants display a white belt of variable width and
position across the body, that is, belted, belted2, dreher, patch,
piebald, rump-white, and splotch (Silvers 1979; http://www.
informatics.jax.org/). Most of these have been shown to affect
melanocyte development and migration, and many are not viable
in the homozygous condition owing to severe pleiotropic effects.
The dominant white spotting (KIT/W) and Steel loci (encoding the
mast-stem cell growth factor receptor and its ligand, respectively)
are to date the loci best characterized at the molecular level (Cope-
land et al. 1990; Nagle et al. 1994; Kluppel et al. 1997). Other
examples are the mutants lethal spotting, piebald, and piebald-
lethal at the loci encoding endothelin type 3 (Edn3) and its recep-
tor, Ednrb (Baynash et al. 1994; Hosoda et al. 1994). Both the
genes for endothelins (Edn1, 2, 3) and their receptors (Ednra and
Ednrb) have multiple functions in early development, including
the proliferation of neural crest derivatives (Reid et al. 1996).
Owing to the large number of potential candidate genes that
may cause a belted phenotype, we decided to first map Belt to a
chromosomal region. We report the result of a genome scan, using
a commercial crossbred population segregating for the white belt
trait. We show that the white belt is controlled by a dominant allele
showing no recombination with the Dominant white locus.
Materials and methods
Animals and coat color phenotypes.
Purebred Hampshire animals
(white belt on black background; Be/Be) had been crossed with purebred
Pietrain animals (white with black spots; be/be). A total of 228 offspring
resulting from mating between 11 F
sows and two Pietrain sires were
used. All F
and backcross (BC) animals were scored for coat color phe-
notype. Only the subset of BC animals unambiguously assigned to the
classes “white belt on black background” (Be/be) and “solid black” (be/be)
were used for linkage mapping. Either ear notches or tails were sampled for
DNA samples from the backcross pedigree were pre-
pared from frozen tails or ear notches with the reagents provided in the
Wizard Genomic DNA Kit (Promega). Tissues were cut into small pieces
and incubated at 65°C for 30 min in 600 l of lysis buffer, containing
) and 0.5 mg/ml of proteinase K. Three l of RNase solu-
tion was added, and the samples were incubated for an additional 15 min
at 65°C. After cooling, 200 l of protein precipitation solution was
added, and the samples were centrifuged. The supernatant was collected
Correspondence to: Leif Andersson, Department of Animal Breeding and
Genetics, Swedish University of Agricultural Sciences, Uppsala Biomedi-
cal Center, Box 597, S-751 24, Uppsala, Sweden.
Mammalian Genome 10, 1132–1136 (1999).
© Springer-Verlag New York Inc. 1999