Jump to content
BC Boards
Mark Billadeau

ABCA Meeting: Health & Genetics Summary

Recommended Posts

The results of the ABCA funded Genome Wide Association Study (104 epileptic dogs; 104 healthy dogs; 174,000 marker gene chip) found no region in the genome associated with epilepsy in Border Collies (no test is possible). The genetic component of epilepsy in Border Collies is complex (multiple genes) enough that Dr Mark Neff believes it cannot be selectively bred out of our gene pool. Based upon the study results, my personal recommendation is to not repeat a cross that produced epilepsy but there is no evidence to support removing the sire and dam from the breeding population.

 

 

The test data (already paid for) from the 204 dogs can be "mined" for other genetic factors (behavioral, health, physical, etc); all that is needed is an assignment of each dog for the new factors of interest (i.e. strong eye vs weak eye; wide running vs tight running, etc). We will be working with Dr Neff on a list of possible factors for more Genome Wide Association Studies using the data we have.

 

Dr Neff has previously reported finding a chromosomal region associated with EOD but that the Genome Wide Association Study data could not identify the exact location of the mutation responsible (required to develop an EOD test like the one for CEA). He has 10 likely locations for the mutation. He has proposed a risk test where dogs that have two copies of the normal genes at all 10 sites are not likely to go deaf; dogs that have 2 copies of the mutated genes at all 10 sites are likely to go deaf; and dogs with 1 normal and 1 mutated gene at all 10 sites are likely carriers. Those dogs that have different gene sets at the 10 sites could be used to determine which of the 10 sites has the gene responsible for EOD; this information could lead to the development of a EOD test like the CEA test. Dr Neff is working on a process to provide sample kits (for a suggested donation of about $40) and to collect the required information. He will be interested in testing key dogs (ones from affected lines).

 

 

The clinic study for BCC was completed last year; for no additional cost the researchers have been able to initiate a Genome Wide Association Study from the DNA sample they collected as part of this study. Results should be available later this year.

 

 

This summary was provided at last night's ABCA meeting.

 


Share this post


Link to post
Share on other sites

Our initial Genome Wide Asscoaition Study for epilepsy was condiucted in 2008 with 45 affected dogs and a matching set of normal dogs. At that time the state of the art gene chip tested 28,000 locations across the genome. Due to the improvment in gene chips and several publications about finding genes associated with epilepsy in other breeds a second, larger (more dogs), study in Border Collies was initated with the new chips that test 174,000 locations across the genome.

Share this post


Link to post
Share on other sites

The results of the ABCA funded Genome Wide Association Study (104 epileptic dogs; 104 healthy dogs; 174,000 marker gene chip) found no region in the genome associated with epilepsy in Border Collies (no test is possible). The genetic component of epilepsy in Border Collies is complex (multiple genes) enough that Dr Mark Neff believes it cannot be selectively bred out of our gene pool. Based upon the study results, my personal recommendation is to not repeat a cross that produced epilepsy but there is no evidence to support removing the sire and dam from the breeding population.

 

 

 

But since the heritability (or not) of the condition is currently unidentified, do you not think it would be more responsible not to take the chance of breeding from dogs that have produced epileptic offspring in the past?

 

A sample of 208 dogs doesn't seem very big to me but I'm sure it isn't as simple as that.

Share this post


Link to post
Share on other sites

The sample was 104 epileptic dogs and 104 normal dogs; a large sample of affected dogs. According to Dr Neff, this is larger than any other published study on epilepsy using the 174,000 marker gene chip.

 

For example, a recent report located a region in the Belgan Shepherd genome that was associated with epilepsy in that breed using 40 affected and 44 normal dogs. That study used a gene chip that testing 50,000 markers.

Identification of a Novel Idiopathic Epilepsy Locus in Belgian Shepherd Dogs

 

 

Inheritance of a single gene conferring high risk of developing epilepsy would have been identified by this study. From what I can tell from the study and talking with Dr Neff, the genes responsible for producing epilepsy in our breed are equally likely to be present in dogs that have produced epilepsy as those that have not. The science based recommendations do not include excluding breeding dogs that have produced epilepsy, only exclude affected dogs and repeating the cross that produced epilepsy.

 

Repeating a cross that produced epilepsy could easily never produce epilepsy again (polygenic disease with 2 alleles per gene and an unknown number of homozygous genes required to produce epilepsy); but in the abundance of caution we are recommending to not repeat the cross.

 

 

Share this post


Link to post
Share on other sites

For example, a recent report located a region in the Belgan Shepherd genome that was associated with epilepsy in that breed using 40 affected and 44 normal dogs. That study used a gene chip that testing 50,000 markers.

Identification of a Novel Idiopathic Epilepsy Locus in Belgian Shepherd Dogs

 

Could results in other breed help guide research? i.e. provide a more refined testable hypothesis, it's after all the same species but perhaps gene differentiation of the disease is clearer in another breed?

 

I assume they've tested the hypothesis that all dogs may not have 'the same' epilepsy? i.e. that there are not one gene (set of genes) for epilepsy but perhaps several disting ones? That would make analysis of 'affected' vs 'clear' much harder to do. But I 'assume' still possible probabilisticaly (as long as they've run correlation on various subsets of the population and genes)

Share this post


Link to post
Share on other sites

Dr Neff was aware of this study and it was one reason why we chose to do the recent Border Collie study.

 

Most of the literature indicates the few genes (2 cases that I'm aware of) are specific for that breed. In the Belgan Shepherd study they tested affected and normal dogs from 8 other breeds (Lagotto Romagnolo, Miniature Pinscher, Kromfohrländer, Whippet, Border Terrier, Schipperke, Finnish Spitz and Finnish Lapphund) for the gene found in Belgan Shepherds and showed a tentative association in only 2 of those breeds.

 

I believe the 174,000 makers in the gene chip Dr Neff used for the Border Collies include the 50,000 markers in the gene chip used for the Belgan Shepherd study. If the same gene was present in Border Collies it would have been found by the chip used in our study.

Share this post


Link to post
Share on other sites

Thanks for this update, Mark! Do you know if there is a summary available for sharing somewhere on the internwebs?

Share this post


Link to post
Share on other sites

Thank you, Mark, for the good presentation last night and for all this information.

Share this post


Link to post
Share on other sites

Mark

Thanks so much for getting this information to us right away. It sounds like there is good information coming out to improve understanding of these diseases and help us understand how to make better breeding decisions.

Share this post


Link to post
Share on other sites

Thanks Mark for all the valuable info! Working in a vet clinic with a lot of Border Collie fanatics you can probably imagine how many people have heard about and are reading this post today. I'm looking forward to the official research report. Do you know if it will be made available to the general public and posted in a journal?

 

Thanks!

 

Bethany, Rose, and Loki

Share this post


Link to post
Share on other sites

Mark, does this mean that we can get tentative results for EOD if a sample is submitted? Confused.

Share this post


Link to post
Share on other sites

A region of one chromosome has been identified as the location of the EOD mutation; the exact site of the mutation has not been identified. Dr Neff has 10 sites in the chromosomal region that he feels are likely locations for the exact site of the mutation.

 

 

He believes:

a dog that is homozygous normal at all 10 sites the dog is likely normal for EOD

a dog that is homozygous mutated at all 10 sites the dog is likely affected with EOD

a dog that is heterozygous at all 10 sites the dog is likely a carrier

 

 

These are the only dogs that he will be able to assign a risk assessment for EOD.

 

 

For those dogs that do not have the same results at all 10 sites he will not be able to assign a risk assessment for EOD; however, by following these dogs he hopes to be able to eliminate some of the 10 sites. These are the dogs he is interested in finding a following and are key in finding the exact site of the mutation.

 

 

For example if a dog was found to be normal/normal at 3 sites, normal/mutant and 1 site, and mutant/mutant at 6 sites and then went deaf he would know that 4 of the 10 sites are not the location of the EOD mutation.

 

Until a exact location of the mutation has been found and proven to cause EOD; all test results can only be considered risk assessments.

Share this post


Link to post
Share on other sites

So we can get those results at this time for dogs already in the study?

Share this post


Link to post
Share on other sites

Thanks so much, Mark! It's encouraging to know so much progress has been made with EOD. Bummer about the epilepsy - I had been hoping that the study with Belgian sheepdogs would have provided the key in Border collies as well. These two diseases are such heartbreakers...

Share this post


Link to post
Share on other sites

In a Genome Wide Association Study (GWAS) dogs are segregated into affected and normal (which includes carriers), those locations that are changed in the affected dogs and not changed in normal (and carrier) dogs are associated with the disease. The data from this test cannot be used to look for carriers.

 

 

The GWAS tested very specific locations spread across the entire genome. The 10 sites Dr Neff is proposing fall between the locations tested in the GWAS and will test both alleles (one from each parent) at each site.

 

Share this post


Link to post
Share on other sites

I have had multiple people inquire privately if the samples already submitted and tested as part of the GWAS can be used to identify likely carriers and undiagnosed affected dogs. I have asked Dr Neff to provide an answer to this question and a lay person explanation as to why. I will post his answer when I receive it.

 

 

 

Everyone should understand that a GWAS is like a correlation study (for example: incidence of autism increased with increased rate of vaccination) it does not demonstrate causation. While changes in a chromasomal region are asscoaited with EOD; there is no proof (yet) that those changes cause EOD.

Share this post


Link to post
Share on other sites

You can think of Genome Wide Association Studies this way.

 

Imagine you had reports of car stereos getting stolen out of different cars at different times on different days.

 

So, you plot the locations of all the cars on a map of all city streets.

 

You might find that the cars that got broken into were evenly distributed all over the city (epilepsy), or you might find that all of the cars that got broken into were in two neighborhoods, and within five blocks of each other (EOD).

 

The second one doesn't give you the address of the stereo thieves, but it suggests two areas to start looking in for them.

 

My guess is that GWAS data

 

Pearse

Share this post


Link to post
Share on other sites

What is the recommended approach of a sibling from a litter with pups with epilepsy in regards to breeding? If the sibling is healthy over 4 yrs old, does it have a higher probability of producing an epileptic pup than the general population?

Share this post


Link to post
Share on other sites

They can still develop epilepsy even after 4 years old. (Ask me how I know this. :( )

 

Mark can give you the ABCA recommendations--I'll just say that I would be leery of doing this, though presumably if you bred to a dog completely unrelated to the two who produced the epileptic pup(s) you would reduce your chances of producing epilepsy.

 

I had a well-known handler from the UK once say to me that he would have taken a chance breeding to a dog who was the sibling of an epileptic because he liked the dog that much. But of course many people aren't willing to take such a risk, understandably.

 

J.

Share this post


Link to post
Share on other sites

The ABCA recommendations are to not breed an affected dog or repeat a cross that produced an affected dog.

 

 

 

The inheritance and the genetics of epilepsy are not known; no genetic association was found in affected dogs that was not also present in normal dogs. Based upon this, one could infer there is no more risk breeding an unaffected sibling as there is in breeding any other unaffected dog.

 

 

The key question is at what point in time can we be certain that this sibling is not affected? I don't think there is a definitive answer to when epilepsy is expressed.

Share this post


Link to post
Share on other sites

Yes, Julie, I should have put it at 6yo, although I know that it can happen at any age, only the probability of secondary epilepsy is then much higher.

 

Thank you for your answers. I was so hoping the gene would be found.

Share this post


Link to post
Share on other sites

Our recommendation is to not breed affected dogs or repeat a cross that produced an affected dog. Science studies do not offer better guidance.

 

No studies have been able to identify genetic correlations to affected dogs. I was just reviewing a recently published study that found no conclusive genetic markers correlated to affected dogs. The authors suggested the state of the art gene testing does not have the resolution to identify genes correlated to epilepsy.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×