History of the StudyThe following article was written by Anita M. Oberbauer, Ph.D. and originally published in the American Belgian Tervuren Club's newsletter, the TN
In the June/July 1996 TNT, Dr. Thomas Famula presented a summary of results from a 1986 seizure survey sent to all ABTC members. The impetus behind the ABTC survey at that time was to provide breeders of Belgian Tervuren with information about the heredity of epileptic seizures. The data languished for a number of years, before being sent to Dr. Famula, a quantitative geneticist, for evaluation.
The data of 997 unique Tervuren were first analyzed with the objective to quantify the role of inheritance in onset of seizuring activity. As previously reported in the TNT, the incidence of Tervuren seizures, as determined from the survey data (which included three generation pedigrees), was 17% with a high heritability estimate (Famula et al., 1997).
Heritability estimates provide a basis to determine if genetic selection programs (i.e., concerted selected breeding) can impact the prevalence of a particular trait. Heritability estimates reflect a composite of environmental influences (such as diet, temperature, exercise) and genetic influences. In other words, heritability estimates have predictive value as to the likelihood that improvement or change can be achieved by using specific breeding stock.
To put the heritability estimate of Tervuren seizure activity in perspective, reproductive traits, such as how many puppies will be born for a given dam, are difficult to select for and achieve measurable results from one generation to the next. Those traits, which are highly influenced by the environment, have low heritability values in the range of 0.15 or lower.
In reproductive traits, the environmental influences swamp out any detectable genetic influences. In contrast, growth traits, such as height, have relatively high heritabilities (~0.4) and significant progress (increased height) can be made by breeding animals of the desired height. The genetic influence on expression of a growth trait is relatively great and therefore one can exert more effective selection pressure.
Does any trait have a heritability of 1.0? Or rather, what trait is fully governed by the underlying genes and not influenced by the environment? In mammals, the sex of an animal has a heritability of 1.0; a normal male is a male if he carries the Y chromosome regardless of how well the mother ate or exercised during her pregnancy.
The heritability estimates for traits therefore give an indication of how much of the trait of interest is governed by genetics. Heritability estimates are frequently given as a range due to the imprecision of the data, calculations, underlying assumptions, or environmental influences on a particular population under analysis.
The range of the heritability estimate for seizuring in Belgian Tervuren is 0.65-0.88. In other words, highly heritable.
Estimates in that range indicate that significant progress can be made to reduce the incidence of seizures in Tervuren merely by judicious breeding.
What Can Breeders Do?
However, in the absence of an extensive database describing and characterizing breeding animals on their likelihood of carrying or passing on the seizuring condition, what can concerned breeders do?
The data from the Belgian Tervuren survey was additionally analyzed to directly address this concern and is published (Famula and Oberbauer, 1998). To paraphrase this study, by using statistical sampling methods, the probability of producing seizuring offspring can only be estimated when the criteria available for use in selecting breeding animals is the seizuring status of those parents.
Clearly, this is not the ideal way to go about developing selective breeding programs in an attempt to reduce the incidence of this debilitating disorder.
The Single Gene Theory
Analysis of the Belgian Tervuren data (Famula et al., 1997), although clearly indicating a highly heritable basis to seizures, has ruled out a single locus model of inheritance and a supported a polygenic (multilocus) model.
However, heritability estimates of the magnitude detected for Tervuren indicates a single locus/gene that significantly influences the expression of the trait. That is the situation with the Tervuren and seizures.
The data was analyzed by six different methods to assess the presence of a major gene that plays an important, though not exclusive, role in the incidence of epilepsy. While no single one of these tests can be considered conclusive proof that there is a single gene of major effect on seizures, the fact that all six independent tests point to the existence of a major gene, strongly indicates that one major gene does exist.
Seizuring activity due to this major gene appears to be an autosomal recessive.
We have provided an analogy to illustrate how a single gene, though not fully determinant in development of the trait, can significantly influence the outcome.
Developing a Marker
Loci of large effect (major genes) have been utilized in other species for the development of genetic markers to assist breeders in selecting breeding stock that are less likely to pass on the genetic disorder. We will utilize the rapidly developing canine genome map and polymorphic canine markers to develop a linkage between a marker and the seizure phenotype for the Belgian Tervuren.
The most difficult aspect of developing a marker linked to some genetic disorder is identifying the mapping population. A mapping population refers to the individuals that are screened for an association between a genetic marker and the disorder.
The mapping population is the most critical consideration. Ideally, we wish to minimize the background noise due to natural variation among individual dogs that is totally unrelated to seizuring. Every animal different from every other animal. While this difference is to be expected, we do not want to get distracted by differences that have nothingto do with seizures.
For instance, if we evaluated every Belgian Tervuren that its owner classified as being epileptic, it would take years to wade through all the natural variations that have just accumulated by chance (ear set high, ear set low, coat semi-coarse, coat soft, coat a bit shorter than other Tervs, and so forth) in our search for the specific variation linked to seizures. We wish to do a directed search on a sub-population of highly related dogs as they will have less natural variation than the entire population of Tervuren.
Once that marker linkage to seizuring is established in the subpopulation, its general applicability to the entire Tervuren population will be tested.
So what is needed?
We need those subpopulations (families).
And that is where the ABTC membership can help! We need DNA from families of Belgian Tervuren known to be affected by seizures, not just dogs that are classified as epileptic.
What is the definition of a family?
- dog or bitch known to pass on seizuring activity
- offspring from these dogs or bitches
- dogs and bitches these dogs were bred to (even if the breeding did not produce any seizuring offspring)
- parents of either the dog or bitch
- siblings of either the dog or bitch
- repeat breedings of dogs or bitches that have produced seizures, even if no seizures were produced by the repeat breeding dogs or bitches
We need to know the seizuring status of dogs submitted and also, the parents and offspring (as well as can be defined). The seizuring documentation is critical to assign phenotype. Veterinary records substantiating seizures are great.
This study is not limited to dogs within the United States. We encourage breeders from Canada and Europe to contribute to the studies with families afflicted with this genetic disorder. This disease is not limited to Tervuren population in the USA.
How is the DNA collected?
DNA is collected from cells obtained by buccal swabs (cheek cells that you collect on a special "Q-tip" rolled on the inside of a dog?s cheek)
Can you help?
Contact Dr. Anita Oberbauer via email below.
We realize that this is a sensitive topic and some of you may be willing to participate, but only under confidentiality. We have arranged for that also. The previous ABTC survey was collected with the understanding of confidentiality. That goal was achieved in the survey analyses presented here. Computer programmers were screened for their lack of association with purebred dogs. These programmers input the names, pedigrees, and seizure information into a large database. Then the dog names were encrypted such that individual names were replaced with numbers. The analyses for the previous data, as well as what we propose here, only relies upon pedigree relationships; names are irrelevant to the success of the genetic marker search.
If you are interested in participating in the study, but wish to have the dogs remain anonymous, write or E-mail Dr. Thomas Famula. He will send out DNA collection kits which include directions and diagrams for collecting, a return envelope, and questionaire for pedigree and seizure information.
When the samples are returned, the dog?s identity will be encrypted and a numerical identification assigned. That number will be ascribed to that dog?s DNA. The dog?s identity will remain anonymous, yet the pedigree associations will remain intact for the study.
A genetic marker linked to this major gene will significantly contribute to minimizing seizures in our breed. We believe such a search will yield a marker.
If you have any questions regarding any part of this study, please feel free to contact either Dr. Famula or Dr. Oberbauer.
|Dr. Thomas R. Famula||Dr. Anita M. Oberbauer|
|Dept of Animal Science||Dept of Animal Science|
|University of California||University of California|
|Davis, CA 95616||Davis, CA 95616|
Famula, T.R., Oberbauer, A.M., Brown, K.N. 1997. Heritability of epileptic seizures in the Belgian Tervuren. J. Small Animal Practice 38:349-352.
Famula, T.R., Oberbauer, A.M. 1998. Reducing the incidence of epileptic seizures in the Belgian Tervuren through selection. Preventative Veterinary Medicine 33:251-159.