An integrated linkage-radiation hybrid map of the canine genome
Cathryn S. Mellersh,
Donald F. Patterson,
Elaine A. Ostrander,
Clinical Research Division, Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue N., D4-100, P.O. Box 19024,
Seattle, Washington 98109-1024, USA
UPR 41 CNRS, Recombinaisons Ge´ne´tiques, Faculte´deMe´decine, 2 Avenue du Professeur Le´on Bernard, 35043 Rennes Ce´dex, France
Section of Medical Genetics and Center for Comparative Medical Genetics, University of Pennsylvania School of Veterinary Medicine,
Philadelphia, Pennsylvania 19104, USA
Received: 10 June 1999 / Accepted: 23 September 1999
Abstract. Purebred dogs are a unique resource for dissecting the
molecular basis of simple and complex genetic diseases and traits.
As a result of strong selection for physical and behavioral charac-
teristics among the 300 established breeds, modern dogs are char-
acterized by high levels of interbreed variation, complemented by
significant intrabreed homogeneity. A high-resolution map of the
canine genome is necessary to exploit the mapping power of this
unusual resource. We describe here the integration of an expanded
canine radiation hybrid map, comprised of 600 markers, with the
latest linkage map of 341 markers, to generate a map of 724
markers—the densest map of the canine genome described to date.
Through the inclusion of 217 markers on both the linkage and RH
maps, the 77 RH groups are reduced to 44 syntenic groups, thus
providing comprehensive coverage of most of the canine genome.
Genetic analysis of the domestic dog offers a unique opportunity to
unravel the molecular basis of vertebrate development, physiol-
ogy, behavior, and disease. Over the last several centuries, breed-
ing programs aimed at developing lines of purebred animals with
specific physical and behavioral traits have led to the segregation
of over 300 recognized breeds with pronounced differences in size,
shape, and behavior. Rigid morphological conformity within
breeds has been achieved, in part, by crossing closely related in-
dividuals. In the absence of strong selection against genetic dis-
ease, this has led to a high frequency of autosomal recessive dis-
eases, many of them breed-specific (Patterson 1999). Repeated
breeding of favored sires and population bottlenecks have further
contributed to reducing genetic diversity within many breeds.
Thus, a disease shared by individual dogs of the same breed is
probably caused by an identical ancestral mutation. Over 350 dis-
tinct canine genetic diseases have been described, and many of
these, including congenital heart defects, inherited forms of blind-
ness and deafness, motor neuron disease, and several forms of
cancer have clinical and pathologic characteristics that are similar
or identical to human diseases and are thus hypothesized to be
caused by mutations in homologous genes.
We, and others, have suggested that genetic studies of canine
traits could provide ready access to genes underlying mammalian
behavior, morphology, and disease (Scott and Fuller 1965; Patter-
son et al. 1982, 1988; Ostrander and Giniger 1997; Galibert et al.
1998). Other mammalian genetic systems, such as the mouse, do
not exhibit many of the traits or the degree of interbreed variation
seen in the dog. Moreover, the problems of locus heterogeneity
and modified expressivity, which confound human linkage studies,
could probably circumvented by taking advantage of the strong
founder effects and minimal diversity observed in dog breeds aris-
ing rapidly from small numbers of individuals. Thus, breeds of dog
serve as genetic isolates analogous to the isolated Finnish and
Icelandic populations that were used to identify several human
genes. The relatively short generation time of dogs and the possi-
bility of planned matings means that many complex traits, such as
those associated with breed-specific behaviors or morphologic
variation, could be easily mapped in dogs.
The development of canine markers and the ordering of those
markers into a genetic linkage map over the past 2 years has
enabled the identification of linkage between several disease genes
and anonymous polymorphic markers (Acland et al. 1998, 1999;
Blazej et al. 1998; Yuzbasiyan-Gurkan et al. 1997). Full exploita-
tion of the unique resource of canine genetic variation, however,
will require alignment of the canine genome map with that of other
well-studied mammals. Our goal, therefore, is to integrate the ca-
nine linkage map of anonymous polymorphic microsatellites with
a ‘gene map’ of highly conserved loci, and to establish at high
resolution the syntenic relationship between the dog and other
We describe an expanded canine radiation hybrid (RH) map
and its integration with the third-generation canine linkage map
(Werner et al. 1999a). The linkage map comprises 341 markers
(Werner et al. 1999a), and the RH map comprises 600 markers, an
increase of 50% over the first canine RH map (Priat et al. 1998).
The RH and linkage maps were integrated through the duplicate
typing of 217 markers positioned on both maps. This process
associated virtually all the RH groups with specific canine chro-
mosomes or linkage groups, and generated an integrated map of
724 unique markers. Approximately two-thirds of the markers are
polymorphic microsatellites and one-third are genes; thus, the map
is suitable for both linkage analysis and comparative mapping
studies. The integrated map provides the means for candidate gene
studies to begin for initial linkages deduced from the linkage map
by selecting appropriate candidate genes from the corresponding
regions of the well-developed mouse and human maps.
Materials and methods
Generation of the canine radiation hybrid cell line.
and characterization of the canine-rodent hybrid cell panel (RHDF5000)
has been described previously (Priat et al. 1998; Vignaux et al. 1999)
Availability of DNA from all cell lines and information may be obtained
through the web site http://www-recomgen.univ-rennes1.fr/doggy.html.
Authors who contributed equally to this work.
Correspondence to: E.A. Ostrander or F. Galibert
Mammalian Genome 11, 120–130 (2000).
© Springer-Verlag New York Inc. 2000