Introduction
We are entering an era when DNA testing will be possible for numerous traits in animals. Currently, many tests are available for genetic diseases and more are being developed, but other characteristics like coat colour, temperament, behaviour, susceptibility to hip dysplasia, cataracts, cancer, parasites or contagious disease, will become available based on the same technology.
DNA testing for genetic disease can not only confirm a diagnosis of a condition that is difficult to identify, but it can also be used to detect carrier status of genetic diseases--that is animals that show no outward signs but carry one defective copy of a gene for a recessive disorder. Such tests allow breeders to eliminate these diseases from the breed by testing and selective breeding. One of the first such tests available on dogs was copper toxicosis in Bedlington terriers, but now there is a long list of tests available in different breeds. The list of tests is long (e.g., Brooks and Sargan, 2001, Galibert et al, 2001) and most tests are specific to one or a few breeds, so the best way to identify available tests is a web search.
The Origins of Genetic Disease
Genetic diseases will always be a problem in domestic animals because of the breeding structure of extensive use of champion animals and subsequent matings between descendents (inbreeding). Any recessive genetic defect carried by that champion animal, (we all have 5 or so defective genes but the presence of another normal copy means there are no adverse consequences), will show up when descendents inherit both copies of this gene passed down through the pedigree from him. The consequence of this homozygous state (both copies the same--identical-by-descent) for the defective copy of the gene (the disease allele) is genetic disease. So, most genetic diseases in such breeding structures can be traced back to a single individual in which the mutation occurred or inherited the mutation and made it common in the breed.
For example, research in my lab developed a test for ceroid lipofuscinosis (CL) in Border collies (Melville et al., 2005). CL is a lysosomal storage disease with nerve degeneration from around 12 months and results in behavioural changes, ataxia, tremors and eventual death by around 2-3 years. There is no treatment, and few drugs that can alleviate symptoms. In the Border collie show dogs in Australia, the disease was alarmingly common. DNA testing showed that around 3% of animals carried the defective copy of the CL gene. The source of the gene defect could be traced back to a single individual from the 1950s who was a common ancestor to all known cases. Extensive use in breeding of champion dogs that were carriers of the defect made it common in the population. Inbreeding, mating two descendents from that champion dog results in a proportion of affected puppies (for recessive disorders one quarter are expected to be affected in any carrier by carrier mating). DNA testing has allowed diagnosis of this devastating disease, that could only be identified previously by pathology on brain biopsy in grown dogs. Now it can be done in puppies at birth using a non-invasive cheek swabs. More importantly, it has been used to detect carriers of the genetic defect, guided breeders in their choice of matings so that affected litters are not born, and the genetic defect can gradually be bred out.
Another consequence of breeding structure in domestic animals is that these recessive genetic defects will be unique to a breed or a small group related of breeds. CL occurs in numerous dog breeds, e.g., Border collie, English setter, American bulldog, but the defect is different in each breed. In these cases, even the genes involved in the disease are different (Melville et al, 2005; Katz et al 2005, Awano et al 2006) but in other cases, it may different mutations in the same gene. Hence, a DNA test designed for one breed will probably not be useful in another breed unless the mutation predates the development of the breeds, e.g., collie eye anomaly (CEA/CH) is found in all collie breeds and is caused by the same large deletion of DNA on dog chromosome CFA37 (Heidi Parker, FHCRC, at ISAG 2006; Graves 2006) for which testings is available through Optigen.
How is DNA Testing Done?
DNA testing in animals uses the same forensic techniques used in humans for identification and paternity testing. Testing can be done for specific DNA changes that result in genetic disease or animal identification and paternity. Both types of test use the polymerase chain reaction (PCR) to amplify a very small amount of DNA to a quantity that can be analysed. The starting material is usually either blood or mouth swabs, but a tiny amount of any tissue can be used. DNA is extracted and a one or two microlitres is used as a template for PCR. The reaction contains an enzyme to copy the DNA (Taq polymerase), the building blocks of DNA (nucleotides), buffers and oligonucleotide primers, which target the DNA to be amplified. Primers allow just a small piece (<1000 bases) of the 2,500,000,000 base genome to be amplified, which is the key a successful test. The PCR goes through about 30 rounds of exponential amplification. The DNA content of the PCR product then has to be checked. For identification or paternity this is by the size of the products produced in microsatellite repeat regions. For disease testing, this is by identifying a change in the DNA bases or a missing piece of DNA. DNA sequencing of the PCR product will identify the changes but other methods are often used to check for specific changes. Most testing requires access to large expensive automated DNA sequencers to detect fluorescently labeled PCR products separated by size.
If the mutation results in a difference in the DNA, it may be identified by enzymes that cleave specific DNA sequences (restriction enzymes). The DNA carrying the mutation will give a different pattern of cuts to the DNA to the normal DNA and this can be identified by gel electrophoresis, e.g., CL testing in Border collies at the CLN5 gene. Other methods specifically amplify the DNA with the mutation or the normal DNA, in a competitive PCR. When the primers are labeled with different dyes, one colour PCR product will indicate the presence of the mutated copy of the gene, while the other represents the presence of the normal copy of the gene. Both affecteds and carriers can be identified from the colours of the products.
How are DNA Tests for Genetic Diseases Developed?
To develop a diagnostic DNA test samples have to be collected from a number of families with the disease. Genetic markers, like the microsatellites used in paternity testing, are analysed, and an association is sought between the inheritance of the marker and the inheritance of the disease gene (linkage analysis). If there are candidates for the disease gene known from model organisms like humans, which have a wealth of clinical and genetic data, then they can be tested by using markers from the same region as the gene in the dog genome. If there are no known candidates for the disease gene, then the whole genome can be scanned by testing 300 to 400 markers spread evenly over all of the dog chromosomes. Linkage analysis will reveal markers close to the disease gene. Once a disease gene is located, then the DNA of affected animals has to be sequenced and compared to those of healthy dogs to identify the mutation that is causing the disease.
This process can be time consuming and expensive. However, the completion of the dog genome (http://www.ncbi.nlm.nih.gov/genome/guide/dog/index.html and Lindblad-Toh et al, 2005) has greatly facilitated this type of research. It is now much easier to go from a group of sick animals in a breed with no known cause to a DNA test. For example, it took 12 years to identify the cause of CL in Border collies before the dog genome project and only 2 years to identify the gene involved in a hereditary neutropenia (Trapped Neutrophil Syndrome) after.
Disease gene research requires samples from numerous pedigrees with affected offspring, their siblings, their parents and grandparents and relationship to other cases of the disease. Five or more such families are usually required to get enough data for statistically significant results. Since such samples are difficult to collect it is advisable to store DNA from any pedigrees with unusual clinical conditions that be helped by disease gene research and DNA testing. Such DNA banking can be simply and cheaply achieved. Blood spotted onto FTA cards will preserve the DNA for more than 10 years and the cards are easily stored in a dry place such as a filing cabinet until they are needed for research.
The Future Development of Genetic Tests
Already the information from the dog genome has been exploited to develop new tools for disease gene identification. Affymetrix microarray chips that can examine 26,000 single nucleotide polymorphisms (SNPs) in dogs have been developed by the Broad Institute (Lindblad-Toh et al, 2005, Graves 2006) to look at genes involved in cancer in dogs http://www.broad.mit.edu/mammals/dog/donate.html and other diseases. These contain much of the common genetic variation found in dogs. The SNPs are spread throughout the dog genome and can be used to try to locate any disease gene. Just 10 to 20 affected dogs and a similar number of controls are required for this approach. If both copies of the disease gene are inherited from a common ancestor all the genes in that region of the chromosome will be homozygous (i.e., two identical copies). Scanning the whole genome will identify regions of homozygosity that are the likely position of the disease gene. Still the precise gene involved in the disease and the mutation in that gene need to be identified by DNA sequencing of the affected and controls. A project like this can be done in a few months once the samples are available.
The microarray slides are not generally available yet but researchers like myself would like to apply them to our research, e.g., we are looking for the gene that causes ataxia in working kelpies. Such a high-throughput approach, although it requires substantial research funds would considerably shorten the length of the research project.
Conclusion
The era of DNA testing is upon us. Whether it is testing for coat colour in dogs, as available through Vetgen, of coloured points in cats (Lyons et al 2005) or for life threatening genetic diseases the methods are now readily available and an increasing number of tests being made available. The most valuable resource in this type of research is the clinical specimens. Research cannot be done without samples from affected dogs and their relatives. Vets can get involved in this type of research by preserving DNA from interesting cases. This is as simple as spotting some blood onto an FTA blotting paper card and filing it away in the draw. Now is the time to start this DNA banking from clinical cases, and also important breeding animals in each breed so that when the research comes to genetic diseases in that breed (they all have something and new ones will appear all the time while the current breeding structure remains) the lines carrying the genetic defect can be easily identified.
References
1. Awano T et al. A mutation in the cathepsin D gene (CTSD) in American Bulldogs with neuronal ceroid lipofuscinosis. Mol Genet Metab. 87:341-348, 2006
2. Brooks M, Sargan DR. Genetics aspects of diseases in dogs. In The Genetics of the Dog. A Ruvinsky, J Sampson eds. CABI Publishing, Wallington, pp191-266, 2001.
3. Graves K, ISAG Conference 2006, Porto Seguro, Brazil. Dog Genome Mapping Workshop http://www.isag.org.uk/ISAG/all/ISAG2006_DogMapping.pdf
4. Galibert F et al. Canine model in medical genetics. In The Genetics of the Dog. A Ruvinsky and J Sampson eds. CABI Publishing, Wallington, pp505-520, 2001.
5. Katz ML et al. A mutation in the CLN8 gene in English Setter dogs with neuronal ceroid-lipofuscinosis. Biochem Biophys Res Commun. 327: 541-547, 2005.
6. Lindblad-Toh et al. Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature 438, 803-819, 2005
7. Lyons LA et al Tyrosinase mutations associated with Siamese and Burmese patterns in the domestic cat (Felis catus) Animal Genetics 36: 119-126, 2005. doi:10.1111/j.1365-2052.2005.01253.x
8. MelvilleSA et al A mutation in canine CLN5 causes Neuronal Ceroid Lipofuscinosis in Border Collie dogs. Genomics 86: 287-294, 2005
9. VetGen: Veterinary Genetic Services www.vetgen.com/canine-coat-color.html