Genetic Research Strategies: The Example of Canine Epilepsy

Barbara G. Licht, Ph.D., Mark H. Licht, Ph.D., and Kathleen M. Harper, DVM, Ph.D.
Department of Psychology, Florida State University
Tallahassee, FL 32306-1270


The first step in understanding a genetic disease is determining the mode of inheritance. Geneticists and statisticians examine the patterns of inheritance for the disease to determine if the disease is influenced by several genes or primarily by one gene. They also try to determine if the disease is recessive (both parents must be carriers in order for an offspring to have the disease) or if it is dominant (only one parent needs to be a carrier for an offspring to have the disease).

To determine the mode of inheritance, information is needed on which individual dogs have the disease and which of their relatives did and did not inherit the disease. In most studies, these relatives include littermates, parents, grandparents, and offspring. However, if the pattern of inheritance is complex, as we believe it is for epilepsy, information also must be collected on other relatives, such as aunts, uncles, and cousins. Sometimes, the people collecting these data are scientists, sometimes they are dedicated dog lovers, and sometimes they are both at the same time. What is important to know is that this is a long tedious, but crucial process. What is even more important to know is that it is the breeders and owners that possess the very valuable information that is needed for this research. All the scientists in the world can not find the genes that are responsible for diseases in dogs without the information possessed by owners and breeders.

With respect to our research on the mode of inheritance for primary epilepsy in Poodles, I will begin with a fairly standard line that you have heard from many other researchers. That is, on a general level, we have made considerable progress, but we have a long way to go. Below, I will be more specific. I have joked before that I am not capable of giving a short answer to any question. However, I think it is important to explain the reasons for our procedures as well as our findings.

For each phase of our research, I will describe our goals, the importance of these goals, and the degree to which we have reached our goals. This report will include Standard Poodles (SPs) only because we are focusing on Standards first. Later, we also will focus on Miniatures and Toys. Importantly, we welcome new participants who own any variety. The data we collect on Miniatures and Toys at this time will give us a "head start" for when we focus on those varieties.

Phase 1 Goals

During Phase 1, our goal was to collect short "preliminary" questionnaires from ALL owners and breeders of Poodles (all varieties), even if they never had problems with seizures. The purposes of the preliminary questionnaire are:
  1. To identify SPs with seizures and non-seizing SPs who are related to those with seizures,
  2. To obtain a rough estimate of the prevalence of primary epilepsy in the breed.
  3. To identify potential participants for later research which will focus on environmental determinants of seizures.

Importance of Phase 1 Goals

It should be pretty clear why we want to identify Poodles with seizures. After all, how can we study seizures in Poodles without a large and representative sample of Poodles that have had seizures? Many of you also understand why we need to identify NON-seizing SPs who are related to SPs with seizures. (This is explained below.) However, it is least obvious why we wish to receive questionnaires from those who own Poodles that are unrelated to ones with seizures. First, in order to obtain a more accurate estimate of the prevalence of primary epilepsy in Poodles, we also need responses from those who have not had problems with seizures. Otherwise, we might overestimate the prevalence of epilepsy. Although it is impossible to avoid this problem entirely, we are making a strong effort to minimize it by repeatedly encouraging broad participation.

It also is important to obtain questionnaires from those who have not have problems with seizures because for our later research on the environmental determinants of seizures (e.g., vaccines, pesticides, etc.), we will need to compare the environments of Poodles with and without seizures to see if their environments differ. Last, and importantly, some individuals who own Poodles who are related to those with seizures may not realize it. Thus, if those owners are already in our database, we will not have to spend time and money tracking down their address and phone.

Progress in Phase 1

Phase 1 has been successful. However, it is NOT too late to become a participant. Thus far, we are pleased to report that more than 250 owners/breeders of SPs have filled out our preliminary questionnaire and have indicated a willingness to provide further information. Our respondents own an average (mean) of approximately 4 SPs, with a range of 1 SP to 25 SPs. The most common number owned was 1; and the higher numbers reflect breeders who had a litter on the ground at the time they filled out the questionnaire. Our Phase 1 findings include information on almost 1,000 SPs that are spread across a wide range of ages.

Thus far, we have identified owners of 87 SPs with seizures. Additionally, 24 respondents reported that a current SP has produced at least one offspring that has had seizures, and 31 reported that they know of a parent or grandparent of a current SP that has had seizures. Thus, Phase 1 data helped us identify a significant number of SPs with seizures and a significant number of relatives.

As an aside, conducting research on a disease puts one in a strange psychological position. Every time I receive information on a new Poodle with seizures, I am deeply saddened by the suffering of both the owner and Poodle. Because one of our own SPs has epilepsy, I know exactly what the owner is feeling, and it is heartbreaking. However, at the same time, I know that every time we identify another potential participant, we get one step closer to our goal of determining the mode of inheritance for primary epilepsy in SPs. This, in turn, can bring us closer to our long-term goal of identifying the gene(s) that influence primary epilepsy in all varieties of Poodles and other affected breeds.

Phase 2 Goals

During Phase 2 of our research, we are requesting a photocopy of the pedigree and AKC or Canadian Kennel Club registration certificate for all SPs with seizures (regardless of diagnosis) and all SPs who have a relative with seizures. Importantly, we need information on as many relatives as possible. This includes aunts, uncles, cousins, great aunts, etc., as well as very close relatives, such as littermates, parents, grandparents, and offspring.

Importance of Phase 2 Goals

As was true for Phase 1, most people understand why we need to examine the pedigrees of SPs with seizures. However, it may not be clear why it also is important to examine the pedigrees of non-seizing SPs who are related to those with seizures. The reason is that it is not possible to determine the mode of inheritance for a disorder unless we know which relatives did and did not inherit the same disorder.

As a simple (and hopefully clear) illustration, consider the hypothesis that a certain kind of epilepsy in SPs is a simple autosomal dominant disorder. "Autosomal" means that the gene responsible for the disorder is not carried on the sex chromosomes. Rather, it is carried on one of the autosome pairs. This means that males and females have an equal chance of inheriting the disorder.

To understand the term "dominant," I need to define what an "allele" is. For each gene, there may be 2 or more forms of that gene. Each of these different forms of the same gene is referred to as a different "allele." For example, the human gene that determines blood type comes in 3 forms: an "A" allele, a "B" allele, and an "O" allele. Now consider a gene that may be influencing whether a dog has seizures. To be fashionable, we will name the gene pep1, which stands for Primary Epilepsy in Poodles gene #1. There may be one "normal" allele for pep1, which conveys the necessary information for normal functioning of brain cells. And there may be one "abnormal" allele for pep1 that tells the brain cells to fire abnormally and uncontrollably. This may manifest clinically (in the phenotype) as grand mal seizures. There even can be another "abnormal" allele for pep1 that tells the brain cells to fire abnormally, but it may convey the information for a different type of seizure, perhaps one involving only the legs. But for now, let's just consider 2 possible alleles (one normal and one abnormal) for our hypothetical gene pep1. Each SP will inherit one pep1 gene from their sire and one pep1 gene from their dam. Thus, depending on the genetic makeup of the sire and dam, an SP could inherit zero, one, or two abnormal alleles for pep1.

If the disorder is dominant, this means that an offspring can inherit the disorder as long as one parent has one abnormal allele for pep1. That is, "dominant" means that in order to have the disorder, an individual needs only one abnormal allele for that gene. Inheriting a normal allele from the other parent will not compensate. Now suppose that an individual who is affected with the disorder (that is, he or she presumably has one normal allele and one abnormal allele for pep1) mates with an individual that is not affected with the disorder (both the alleles for pep1 are normal). On average, 50% of their offspring will inherit the disorder. Since the unaffected parent has only normal alleles for pep1, inheriting the disorder from this mating is like flipping a coin. On average, half of the offspring will get the affected parent's normal allele and half the offspring will get the affected parent's abnormal allele. However, the 50% rule only holds if you flip the coin many times. For example, you may not get exactly 1 head and 1 tail on your first 2 coin flips. You easily could get 2 heads and zero tails, even if the coin was perfectly balanced. So while most litters from one affected and one unaffected parent will have roughly 50% affected, some litters can have zero affected and some can have 100% affected.

I tried to keep this example simple -- honestly! But, just for thoroughness, I should add that the picture can be complicated by incomplete penetrance. "Incomplete penetrance" refers to the possibility that some individuals who inherit the abnormal allele will be lucky and never get the disorder. (In other words, the abnormal allele may not always penetrate into the phenotype.) Another complicating factor is the possibility that an individual who is affected with a dominant disorder may actually have 2 abnormal alleles for that gene instead of just one. This can occur if both the parents were affected. Although this is not necessary to produce an offspring with a dominant disorder, it can occur.

Now, suppose you want to test the truth of the hypothesis that the disorder is autosomal dominant. To do this, one tries to identify matings of one affected parent and one unaffected parent. To determine if the disorder is autosomal dominant, you need to see if this type of mating produces an average of 50% affected offspring. Thus, one must find out from the breeders of these litters how many offspring were in each litter. Then one must contact the owners of each littermate to determine whether their dog/bitch ever had seizures. Without this information, it will not be possible to determine whether matings of one affected and one unaffected parent produce an average of 50% affected offspring.

Hopefully, the example I presented illustrates the importance of obtaining information on SPs who are related to those with seizures. In this example, information on the mates of affected SPs and all their offspring were used to help test the hypothesis that the disorder was autosomal dominant. Information on the parents, grandparents, and littermates of an affected individual also are needed to test various hypotheses. If the mode of inheritance is complex, as we expect is true, it also will be necessary to include more distant relatives, such as aunts, uncles, cousins, great aunts, etc. Hopefully, this example also has illustrates how much time, energy, and detective work goes into computing one simple percentage.

My choice of an autosomal dominant disorder for this example may have surprised some of you because we usually expect disorders/diseases to be recessive rather than dominant. Indeed, there are many who believe that primary epilepsy in dogs is recessive. However, it is important to also test the possibility that at least some "subtypes" of primary epilepsy in dogs are dominant, perhaps with incomplete penetrance. Research on human epilepsy has shown that some subtypes of "primary" epilepsy are dominant with incomplete penetrance. In contrast, some human "secondary" epilepsies have been shown to be recessive. Unlike primary epilepsies, these secondary epilepsies involve neurological deterioration and mental retardation in addition to seizures.

Progress in Phase 2

Thus far, we have received pedigrees for 48 SPs with seizures and 31 pedigrees of non-seizing SPs who are related to those with seizures. We did not request pedigrees from owners/breeders whose SPs were not known to seize or to be related to an SP with seizures. However, a number of these owners/breeders graciously volunteered these pedigrees.

Although we are short of our goal for Phase 2, we are pleased with the progress we have made thus far, and we are optimistic that we eventually will reach our goal for Phase 2. Our goal for this phase is to receive pedigrees on the majority of SPs that were identified in Phase I as either having seizures or being related to an SP with seizures. As I said in a much earlier post (on breeders and seizures), we realize that sometimes people who really do want to help somehow get busy with other things. Thus, our Phase 2 packet of long instructions gets put aside. In the future, we will send out reminder notices in the hopes that we can prompt these participants to send us their pedigrees. Additionally, some of our Phase 1 participants have not yet been mailed the request for sending pedigrees. These will be mailed shortly. We also will mail a thank you and update to everyone who did send us pedigrees and registration certificates.

Preparation for Phase 3

In Phase 3, which is just beginning, we will be collecting information that will help us differentiate between cases of seizures that are due to primary epilepsy and cases of seizures that are symptomatic of another disorder. We also will begin collecting information that will help us determine if there are different subtypes of primary epilepsy.

This article is based on a progress report to members of the Poodle Support Group and was edited for the Web by Dr. John Armstrong, University of Ottawa.

For additional information on the Poodle Epilepsy Project, email

December 24, 1977