Molecular tools to analyse and identify inherited diseases have become quickly available over the last five years. To date DNA polymorphisms covering the entire canine genome can be used by all well-equipped laboratories, to identify markers for any disease. DNA libraries of the canine genome containing large-insert clones of dog DNA are also available, which is required to focus on the smallest possible region linked to the disease, when a linked marker has been identified. When a small chromosomal region has been identified, it is possible to analyse genes of interest in that area. Here the availability of the genome map of other species, like man and mouse, is very important. Comparison of syntenic regions of different species almost always reveals genes shared by all mammalian species. With such methods it is now possible to find any functional gene mutation, even when it is an yet unknown gene of which the function has not yet been identified. The internet has become as important as the lab in recent years.

A good example of this approach is the discovery of the deletion in the Murr-1 gene in Bedlington terriers, which causes copper toxicosis in the liver. This was one of the first known inherited liver diseases in dogs, affecting more than 40% of the population in many countries some 25 years ago. We have analysed the population in the Netherlands over the past 25 years, taking liver biopsies for copper staining in virtually all dogs. With this information the breeders could avoid to use affected dogs and also many obligatory carriers. Nevertheless, in this long period they could not do better than reduce the incidence in the Dutch population to about 15%. Three years ago, Vilma Yuzbasiyan-Gurkan published the discovery of a microsatellite DNA marker linked to the disease. With all the DNA material available from our previous studies in the breed, we could show that the marker was indeed tightly linked to the disease, and that at least in the population in the Netherlands, Belgium and Western Germany, there was no loss of linkage in any family. This made the DNA marker suitable as diagnostic test with an accuracy of at least 96% in our population. With the availability of this test the breeders could choose not to produce affected dogs any more, and so they did. In three years time there is no affected dog born in this population, and the disease has extinguished in a few years time. Recently, in our further studies, we could identify the Murr-1 gene in which a deletion causes the disease. In fact, the DNA marker was localised in an intron of the gene, which explains the absence of loss of linkage. The protein for which the gene encodes, is formed in many different tissues other than the liver, and it remains to be studied why the phenotypical effects are exclusively in the liver.

This approach can be exploited now for analysis of other inherited diseases of the liver as well. Especially simple autosomal recessive diseases can be approached in the above way. The most frequent, and also very invalidating disease, which is therefore one of the fist diseases to study, are the congenital portosystemic shunts. The intrahepatic shunts, which occur in many large dog breeds, have the above mode of inheritance and can thus be approached with the linkage analysis strategy. Extrahepatic shunts seem to be inherited in a polygenic way, which makes it more difficult, but not impossible to find the underlying gene defects.

The easiest type of inherited disease, in which there is an obvious candidate gene, is quite rare in the canine liver. There are several inborn errors of ammonia metabolism due to urea cycle enzyme defects, which could easily be resolved by biochemical analysis to identify the enzyme involved, and then by molecular analysis to identify the underlying mutation. Although scientifically interesting, such diseases are so rare that they have little veterinary relevance.

The most demanding type of genetic analysis in the near future is that of the polygenic diseases. For the liver, chronic hepatitis is a breed-associated disease. This means that certain breeds are predisposed, but it is likely that several genes are involved. Moreover, the presence of predisposition in an individual will only become obvious when it is exposed to environmental causative factors. Such diseases will be the challenge of genetics in the coming years.