Color Implications in Health By Sheila M. Schmutz and Tom G. Berryere University of Saskatchewan, Saskatoon Canada schmutz@sask.usask.ca Coat color has been of considerable interest to dog breeders for many years. In 1957, C.C. Little provided a comprehensive hypothesis with many genes and alleles to explain most of the colors and patterns in dogs, based on breeding records. For the past few years we have studied genes involved in the pigmentation pathway at the DNA level (http://skyway.usask.ca/~schmutz/dogcolors.html). Many of Little's predictions have held true but some have not. We found DNA polymorphisms in several genes in the pigmentation pathway. We used these polymorphisms or markers to map genes named EDNRB and DCT to chromosome number 22, KITLG to chromosome number 15, TYR to chromosome number 21 and TYRP1 to chromosome number 11. These markers have also been used in family studies to follow whether the same variant of the marker always was inherited with a particular coat color, pattern, or associated health problem. We have identified multiple mutations in some genes such as TYRP1 that cause brown color, none of which have ill effects on health. Whereas other coat color genes have multiple effects, including some on health. For example merle, which when homozygous, usually causes deafness and sometimes also serious eye problems. We have conducted a family study using Australian Shepherds and have excluded several genes as the cause of merle: EDNRB, KITLG, MITF, ASIP, TYR, PAX3. We are currently studying Harlequin and merle in Great Danes using the polymorphisms in these same genes. Family and data collection are still ongoing but results thus far are more optimistic. (Harlequin Great Danes are thought to be HhMm, or heterozygous at both the Harl and merle causing genes and so finding which gene causes the Harlequin pattern may also lead us to find the gene that causes merle. It was suggested that "albino" Dobermans may have a mutation in tyrosinase, since that is the cause of many types of albinism in humans and mice. However, in a recently completed study, the entire coding sequence of TYR was normal in such Dobermans (GenBankAY336053) and also in blue and Isabella Doberman Pinschers. These results suggest that the P gene may be the cause of this albinism. We did find that these same Isabella and red Dobermans were both homozygous for the TYRP1 proline deletion mutation indicating both are actually brown in base color. Black Hair Follicular Dysplasia is a disorder that has occurred in my own kennel in Large Munsterlanders. Areas of the coat that are normally black in color are grey at birth and then these weak hairs break and fall out. The underlying darkly pignmented skin is also wrinkled and sometimes pimply. Using this litter we have shown that neither MC1R, DCT, TYRP1, KITLG, nor TYR are the genes causing this disease. Colleagues at the University of Pennsylvania plan to continue studies of this disease and samples were also sent to other collaborators in Germany. Currently, we are also studying MITF in microphthalmicied multipe forms of this gene in the dog (GenBank AY2400952). Alternate start codons and some differential splicing of exon 1 and 6 result in different forms in different tissues. Thereby a mutation in exon 1 may lead to an effect on the eye and not on coat color, whereas a mutation in exon 7 could affect both tissues. We have recently identifed the mutation in MC1R that causes melanistic mask. At amino acid 264, methionine is replaced by valine. Some dogs with melanistic mask have premature greying of the muzzle but this does not seem to be correlated with whether they are homozygous or heterozygous for mask. Some coat patterns such as Harlequin, merle and Irish spotting can make it impossible to see the mask. Likewise dogs that are black or brown or blue do not show their mask against their similar body color. This further confirms that dogs in breeds where mask is part of the standard such as Bullmastiffs and Boxers, the reddish coat colors are due to the agouti alleles and not an ee genotype at MC1R. In the course of this same study we were able to show conclusively that brindle is not caused by an MC1R allele, as Little had predicted . Several dog owners had found that the prediction of an Ebr allele did not fit their breeding data either. We thank the many dog owners who have contributed DNA samples from their dogs or complete litters to our study. In addition we are particularly grateful to C.A. Sharp (Australian Shepherds), J.P. Yousha (Great Danes), and Ione Smith (Doberman Pinschers) who coordinated collection of groups of animals in particular breeds for family studies. Key to Gene Abbreviations: Gene Abbreviation     Gene Name C. C. Little's Locus ASIP agouti signalling protein A locus TYRP1 tryrosinase related protein 1 B locus TYR tyrosinase C locus? MC1R melanocortin 1 receptor E locus DCT dopachrome tautomerase EDNRB endothelin receptor B KITLG KIT ligand MITF microphthalmia transcription factor PAX3 paired box 3 protein