Barbara Skelly MA VetMB PhD CertSAM DipACVIM DipECVIM-CA
Department of Veterinary Medicine
University of Cambridge
Cambridge CB3 0ES
Buccal (cheek) swab packs can be obtained from Bryan McLaughlin at the AHT.
Here is a progress report from Louisa Hayward BSc RVN (Graduate Research Assistant, Canine Genetics Research Group, Animal Health Trust)
What stage are we at with this research?
Scientists at the AHT, alongside Dr Barbara Skelly from the University of Cambridge, have continued to work hard on our challenge to unravel the genetic basis of idiopathic epilepsy in the Keeshond. In our last report we outlined our plans for an in-depth genetic analysis known as a genome-wide association study (whole genome scan). This update summarises the outcome of our investigations thus far and makes suggestions for the best way to take this research forward.
What did the whole genome scan involve?
The whole genome scan carried out earlier this year compared thousands of DNA markers in epileptic Keeshonds (‘cases’) with those in non-epileptic Keeshonds (‘controls’) looking for one or more regions of the genome that were consistently shared between the cases, but that were different in the controls. Analyses of the resulting data were carried out in two stages. The first analysis compared DNA samples from 19 of the most robustly diagnosed epilepsy cases with 29 controls, affording us the best opportunity to identify any signals suggestive of an “association”. In a second analysis we sought to boost the chance of finding a signal by increasing the number of cases to 27. However, although we normally anticipate that a larger number of cases will benefit the analysis, their inclusion introduced a trade-off in that the additional eight samples were from less well-defined cases – either because we were lacking clinical information, or because these dogs did not fit the classic clinical picture of epilepsy quite as closely as others in the study.
What were the results?
If epilepsy in all the cases we included in our analysis had been caused by a single, identical recessive mutation we would have expected the analysis to have identified the region of the genome associated with the condition (and that therefore harboured the mutation). Unfortunately the analysis did not reveal a single region of the DNA that was consistently shared between cases, meaning that it is very unlikely that epilepsy in the Keeshond is caused by a single recessive mutation. At this stage we cannot distinguish between other modes of inheritance, including multiple, different recessive mutations that cause clinically similar forms of epilepsy; dominant mutation(s); or indeed that the condition is genetically complex, being caused by variants at multiple positions in the genome, with or without the interaction of environmental factors.
What happens next?
We need to collect and investigate DNA from more dogs. In our last report we emphasised that the number of samples we were sending away for the first whole genome scan may not be large enough to give us a definitive result. We suggested that sample collection should carry on in anticipation of this being the case, and we have been very pleased to see samples continuing to arrive. Our aim is to collect a similarly-sized set of samples for a further whole genome scan which we can analyse in combination with the first dataset to boost our power to identify region(s) of the genome associated with epilepsy.
One effect that was seen to some degree in our first dataset is known as ‘population stratification’, and this can sometimes inhibit a signal from showing itself as definitively as it might. While some of our epileptic Keeshonds were bred overseas, the majority of our control samples came from unaffected dogs born in the UK. Although we can allow for this to a certain extent in the analysis, and we are aware that the Keeshond breed is a relatively international community, we can strengthen our chance of success by more closely matching the origin of our controls to those of our cases in the next set of samples we investigate.
What can we do to further this research?
We continue to encourage you to send us DNA from any Keeshond with a diagnosis of idiopathic epilepsy. It helps us considerably if you can include as much clinical information as possible with your submission as this helps us to robustly define our cases. If you prefer, we can get in touch directly with the veterinarian treating your Keeshond if you provide us with their contact details and your consent for us to do this. We always welcome samples from older Keeshonds, aged eight years or over, who have never suffered from seizures of any kind – to act as controls in our study. We are especially keen to gather control samples from Keeshonds that are not UK-bred – particularly those based in the USA, Canada and Australia. As always, we are grateful for any health updates you can give us on those dogs for which we currently hold DNA samples. For further information on how to submit a sample, or to send us updated health information, please email Bryan McLaughlin at email@example.com
What is the timeframe for the proposed research?
This important initial phase of the study is funded until mid-March 2014. We can therefore continue to collect DNA from additional cases and controls for this stage of the research until Christmas 2013, which will give us time to gather as many samples as possible but still leaves us enough time to undertake the whole genome scan and analyse the data before the funding expires. After this time we will still welcome further samples as they will be needed for the next stage of the research, so please continue to send them in.
Our grateful thanks go to the Kennel Club Charitable Trust for extending the funding period for this study, giving us the opportunity to build on the hard work that has already been done. We also thank Anji Marfleet and Jane Saunders for their unwavering assistance, together with all those owners and their Keeshonds from around the world who continue to support this project.
Some good news ‘hot off the press’ from the Kennel Club is that under the Kennel Club Assured Breeder Scheme, Keeshonds have had its requirements and/or recommendations adjusted as follows: DNA test for PHPT is now a requirement. This change is effective from January 1st 2013 and Assured Breeders have a period of grace of six months in order to come into line with this new requirement.
Below is some not so goods news which is the latest update on the research being done on our behalf from Louisa Hayward BSc RVN (Graduate Research Assistant, Canine Genetics Research Group, Animal Health Trust)
We have now had the opportunity to analyse the Illumina CanineHD BeadChip genotyping data (“genome scan”). We did this in two stages. The first analysis included 19 cases that were categorised as having the most well-defined phenotypes (clinical signs), matched to 29 unaffected controls. Unfortunately, after adjusting for population stratification, this did not yield any strong signals that might be suggestive of an association. For the second analysis we included a further eight cases in which the diagnosis of idiopathic epilepsy was perhaps less robust – either because the history or workup was less complete, or because the clinical signs were not considered to be quite as “classic”. The inclusion of these dogs unfortunately added no strength to the first analysis.
Obviously this result is disappointing – but perhaps not surprising considering the complex nature of this condition. The fact that we genotyped this number of cases and controls but failed to obtain a strong signal suggests that idiopathic epilepsy in the Keeshond is not inherited in a simple recessive manner.
Unfortunately this means that the only way we can make further progress is to recruit further cases and controls to add to this sample set – perhaps setting a total target of 48 cases and 48 controls. Keeshond owners have continued to send in samples to us, from both affected and unaffected dogs. We might suggest recruiting more control samples from dogs based in countries outside the UK to help mitigate some of the population stratification effects that we could see – our cases were perhaps more internationally distributed than our controls, which were mainly from the UK.
So please, if you have an affected dog or know of one, please get in touch with Bryan Mclaughlin (email firstname.lastname@example.org) for swab kits and full details of what to do.
The same goes for the controls; if you have a fit & healthy keesie over the age of 8 years and want to help us beat this disease by being a control sample – please contact Bryan.
Although we really did hope that we had sufficient data for success, all is not lost. We just need more samples from all of you out there from all 4 corners of the world. As I have said before, we can’t test promises – we need the samples!
I am pleased to be able to give at last an update on the epilepsy research being done on our behalf at the AHT.
Canine Genetics Research Progress Report – Idiopathic Epilepsy in Keeshonds October 2012
Why do we need this research?
For many years, Keeshond owners and breeders across the world have been working with scientists and clinicians towards the goal of uncovering the genetic basis of idiopathic epilepsy in the breed. Dogs are diagnosed with idiopathic epilepsy if they have recurrent seizures for which no cause can be identified. As in several other breeds, this primary form of epilepsy appears to have an inherited, or familial, basis in the Keeshond. Previous research has suggested that the mode of inheritance in the Keeshond might be autosomal recessive, but the pattern remains unclear. Familial epilepsies in the dog have so far proved difficult to characterise at DNA level; the main reason being that multiple genes are likely to be involved. Geneticists usually need to analyse large numbers of DNA samples before they can begin to unravel the mechanisms underlying these complex polygenic conditions. A further complication of idiopathic epilepsy studies is that a diagnosis is not always reached as easily as you might think; seizures in an individual dog can result from a number of underlying causes and, ideally, all possible causes are excluded before a dog can receive a robust diagnosis of idiopathic epilepsy. Despite these and other challenges, Keeshond owners have helped us to build up a sufficiently large collection of DNA samples from epilepsy affected and unaffected dogs that we are at last ready to take the next step in this research.
What does the research involve?
The Animal Health Trust (AHT) is working with Dr Barbara Skelly, a veterinary surgeon at the University of Cambridge in the UK, to investigate idiopathic epilepsy in the Keeshond. Our ultimate aim is to develop a robust DNA test that Keeshond breeders can use as a tool to inform future breeding decisions. We still have a long way to go, but we are pleased to report that we are ready to go ahead with an in-depth genetic analysis known as a “genome-wide association study”. DNA samples will be sent to a specialist laboratory which uses state-of-the-art genetic marker arrays to generate data that we will examine closely here at the AHT. We will compare thousands of DNA markers from dogs affected by epilepsy (“cases”) with those from unaffected dogs (“controls”), looking for one or more regions of the genome that are consistently shared between cases, but not with the controls. Should we find evidence of one or more promising regions, the second stage of our project, known as “fine-mapping”, will seek to identify candidate genes within these target regions that we can explore for DNA sequence errors that might be responsible for epilepsy.
Do we have enough samples?
The enthusiastic participation of Keeshond breeders and owners worldwide means that we are ready to submit samples from around 24 cases, alongside a similar number of controls, to the specialist laboratory for genome-wide scanning. The exact number of samples we send to the laboratory rests on a final quality control step. We have already carried out stringent quality checks on all our samples, as the microarray technology requires exacting levels of DNA quality and quantity. The majority of our samples reach these high standards, but some drop below the threshold for suitability. These samples will not be wasted; they can be used in further experiments in our own laboratory at the AHT should we identify one or more regions of DNA that can be looked at more closely.
When will we get the results?
The genome-wide scan, and analysis of the large volume of data it will generate, is likely to take several weeks to complete. We hope to make preliminary results available to the Keeshond community by the end of January 2013. Our ideal scenario is that the scan yields compelling results that are robust enough for us to proceed to fine-mapping. However, as mentioned earlier, previous attempts to characterise the genetic basis of epilepsy in dogs suggest that this condition is complex.
We should be prepared to acknowledge that the number of samples in this set might to be too small at this stage to pinpoint regions of interest that we can follow up with any confidence. Should this be the case, our only option will be to collect more samples and carry out another genome-wide scan in the future to add to our data.
Can more be done to help this research?
Whether or not we meet with success in this first stage of our study, we encourage you to continue to send us DNA from any Keeshond unfortunate enough to receive a diagnosis of idiopathic epilepsy. It helps us considerably if you can include as much clinical information as possible with your submission. We also welcome samples from older Keeshonds, aged eight years or over, who have never shown any signs of epilepsy – to use as controls in our study. Keeping us updated on changes in the health of any dog we already hold a DNA sample from also helps us tremendously in our continued research. For further information on how to submit a sample, or to send us updated health information, please email Bryan McLaughlin at email@example.com
We would not be entering this exciting phase of our study without your continued support. Our grateful thanks go to all those owners and their Keeshonds who have contributed to this project over the years. Particular mention must be made of both Anji Marfleet and Jane Saunders for their invaluable help; their impressive knowledge of the breed worldwide has been a real asset to our research. We wish to thank Dr Barbara Skelly and the Kennel Club Charitable Trust for arranging funding for this study – without them this work could not proceed.
The research group of Professor Hannes Lohi, working in collaboration with Danish, Swedish and American researchers in an EU-funded project, has made a major breakthrough by identifying a chromosome region associated with the most common form of epilepsy in Belgian Shepherd Dogs. By comparing the genome of dogs with epilepsy and healthy control dogs a gene region in chromosome 37 was discovered, which if homozygous, increases the risk of epilepsy seven-fold.
The identified region has excellent neurological candidate genes for epilepsy and ongoing follow-up research is aimed to identify the specific gene causing epilepsy. Epilepsy genes have not previously been identified in this chromosome region, so the discovery will reveal an entirely new epilepsy gene in dogs and possibly also in humans. The type of epilepsy occurring in Belgian Shepherds is extremely common in also other breeds and thus the discovery may have an impact on the understanding of the epilepsies in different dog breeds.
“There are only few genes in the identified region and I believe that the ongoing analyses will help us to discover the specific epilepsy gene,” says Professor Hannes Lohi who led the research. “This would give us a better understanding of the disease mechanisms and provide us with new diagnostic tools for the disease.”
The Research group of Hannes Lohi has begun an extensive gene-sequencing project in which the entire identified chromosome region will be ‘read through’ with a next-generation sequencing method. By identifying the specific gene mutation an individual’s epilepsy risk could be assessed, although the gene mutation may also be common in dogs that never become symptomatic of epilepsy.
The full report can be read on ScienceDaily
Finding the gene/s responsible for epilepsy is like looking for the proverbial needle in not one, but thousands of haystacks. Hopefully Professor Lohi’s news can narrow it down to one haystack (or less) for Barbara Skelly and the team at the AHT who are researching epilepsy on our behalf. Bryan McLaughlin sent the following update as to how our own Keeshond epilepsy research is progressing.
The aim of this project is to compare the DNA from dogs affected with epilepsy with DNA from unaffected control dogs, in order to identify regions of the genome that are consistently shared between dogs affected with this condition, but not with unaffected dogs. The process is called a Genome-Wide Association Study (GWAS), or more simply, DNA scan and to run this requires decent sized samples of high quality DNA samples.
The canine genetics group at the AHT has received 76 samples in total from the breed since the collection began in April 2011. Of these it appears that 32 are from clinically affected cases, comprising of 13 DNA extractions sent from Richard Goldstein’s lab at Cornell University, 10 blood samples direct from Barbara Skelly, and 9 buccal swab submissions from owners. From the above samples only 23 (7 from Cornell, 7 from Barbara plus the 9 recent buccal swabs) are potentially suitable for initial use in the study, based on the quality and quantity of the DNA provided.
Once region(s) associated with epilepsy have been identified we can carry out additional experiments to ‘zoom in’ on the region(s) identified initially, to hopefully reduce and refine the region of interest. This stage of the project is known as ‘fine-mapping’. Hopefully this is when the remaining samples of DNA from affected dogs can be used. We will then scrutinize the sequence of selected candidate genes within the region to hopefully identify the actual mutation responsible for epilepsy in the Keeshond and develop a DNA test that will be made available as useful tool for breeders. There are no guarantee’s however that a disease responsible mutation will be determined, as epilepsy in its many forms is considered to be a highly complex condition.
If there is anyone out there who has a keesie with epilepsy, who hasn’t yet sent in either a blood sample or buccal swabs to the AHT, then please, please do so.
We would like to thank all owners who have submitted samples and information from their dogs – without either of these we would be unable to make any progress with this project. The AHT always welcomes updated health information about any dogs they have DNA from; in particular, if a dog has a change in clinical status after their DNA has been submitted it is very important for us to be informed of the results.
For those of you with epileptic dogs life can be very stressful, particularly if your dog suffers from clusters of fits that are unpredictable and tend to merge together into long periods of abnormal behaviour. For vets too, this is a difficult condition to manage and therefore the genetic unravelling of epilepsy cannot come quickly enough for vets and owners alike.
After the disappointment of the collaboration with Cornell we entered into a new agreement to work with Cathryn Mellersh and Sally Ricketts at the Animal Health Trust (AHT) and they have been collecting our samples for us co-ordinated by Bryan McLaughlin. Recently, thanks to the Herculean efforts of keeshond breeders and owners and some unrelenting pressure from the breed Health Secretaries, Jane Saunders and Anji Marfleet, the longed for target was reached. We now have 31 samples from dogs with epilepsy and many more control dogs.
What happens next? Now the interesting bit can begin. The AHT are going to perform genome wide association analysis (GWAS) on our samples to see if they can narrow down the bit of DNA we are interested in. As many of you will know, the genome is the full complement of DNA that we have in our cells that is organised into separate units called genes. Genes, and therefore DNA, are made up of combinations of four separate nucleotides. Our dogs with epilepsy will be compared to normal dogs to see whether they differ from them in certain areas of their genome. To look for these differences the AHT will make use of things called SNPs or single nucleotide polymorphisms. These are the result of natural variation across the genome and represent DNA differences between individuals within separate families or lines. SNPs are used to define the haplotype of an animal where the term haplotype refers to the inheritance of a cluster of SNPs.
By examining haplotypes, we can identify patterns of genetic variation that are associated with healthy dogs and those with epilepsy. For instance, if a haplotype is associated with epilepsy, then we can examine stretches of DNA near the SNP cluster to try to identify the gene or genes responsible for causing the disease.
As you can imagine, this process is not quick and the analysis of the data that it yields is complex. I guess what I’m trying to say is ‘don’t hold your breath!’ But, and this is a big but, at least we are on the way to trying to unravel the disease in the way that I referred to in my first paragraph and that means that we have made progress.
One of the last things I have to say is a massive thank you to all of the people who have contributed their time, energy and enthusiasm to this project. I really hope that we will have some results to share with you soon as I know that many people have been waiting a long time for this to come to fruition. Our samples have been collected from all around the world so this has truly been an international effort.
If, by some miracle, there are still some people out there who have an epileptic keeshond that has not been sampled, please do not delay in sending a sample. The AHT are used to working with DNA samples extracted from buccal swabs (cells scraped from the inside of the cheek) so to contribute to this project you do not even have to take your dog to the vet. If you are confident that your dog has a diagnosis of idiopathic epilepsy then we would like to hear from you. If you are not sure then I am happy to discuss your dog’s condition with you. You will be sent a sampling kit from the AHT with instructions how to use it and a short questionnaire about your dog. It is a simple, quick and painless procedure to collect the samples and you could be doing your bit to help future generations of keeshonds. Please act now – we need your epileptic dog!
15 July 2011
I have been asked by many people how the new research into epilepsy is going and, unfortunately, the short answer is that it’s struggling along going nowhere.
Out of a total of nearly 200 DNA swab kits either handed out by myself or requested directly from the AHT only 42 have been returned. Yes, you have read that correctly – 42. The required number of controls has finally been reached (35 keesies aged 8 and over who do not fit) but the number of samples from affected dogs still stands at 12. A response of 20% is pretty appalling by anyone’s standards and means that there are well over 150 swab kits lying around out there unused.
The breed finally has a realistic chance of finding the gene & getting a DNA test for epilepsy; we have the resources & technology and what happens – total apathy. Yes, there have been plenty of promises of samples from affected dogs but you can’t test promises. The swab kits need to be sent in to the AHT sooner rather than later if any progress in the research is to be made.
We need just 12 more samples from affected keesies to start, so to coin a phrase, the ball’s in your court. If we don’t get the samples, the research doesn’t happen – the choice is yours.
21 March 2011
I am pleased to finally report some good news on epilepsy research.
Last year I attended lectures by both Dr Cathryn Mellersh & her colleague Sally Ricketts. Dr Mellersh has a BSc and a PhD in Genetics and leads the Canine Genetics Group at the Animal Health Trust (AHT). She has worked at the AHT for almost nine years during which time she has developed and expanded the Canine Genetics Group considerably. The goal of the majority of studies undertaken by her group is to develop robust DNA tests that breeders can use to determine the genotype of their dogs, with respect to a specific disease-causing mutation, and make sensible breeding decisions that will minimise their risk of producing affected puppies
Dr Barbara Skelly has had to abandon the epilepsy research with Richard Goldstein due to a lack of ability to analyse the data obtained from Cornell & the fact that there is now no DNA left to repeat the analysis. Last week Barbara Skelly went to see Cathryn & Sally at the AHT to try to find a way to resolve the problem of the epilepsy project. And the result is good news – the AHT are keen to collaborate on this project & are prepared to use their DNA collection resource to help with this project. They estimate that we need 24 epileptics & 24 controls to run their Illumina-based screening process. Technology has improved immeasurably over the past few years & they do not need very much DNA from each dog but it is probable that we cannot get away with less than 24 samples. Affected dogs are more of a problem. Given the, presumably, small number of cases in the UK, we are again going to have to ask for epileptic dogs from the U.S and Australia as well as European dogs. This means that we need to embark on another worldwide push for sample collection. As it is by cheek swab, it may be better received but really we need to resample anything that is still alive with epilepsy ASAP.
We are in a strong position in so far as we have enough funds to repeat everything here but a weak position in that all our samples were effectively wasted in our first attempt.
The method used by Cornell uses huge amounts of DNA obtained by blood samples compared to the AHT’s method which requires a fraction of the amount which can be obtained via cheek swabs. This means that sample collection will be through cheek swabbing by the dog owner. Collection packs can be requested directly from the AHT from their sample collection co-ordinator
Bryan McLaughlin (firstname.lastname@example.org). These can be requested on a case-by-case basis. The controls should not be a problem but should be older dogs (over 8 years) who have no history of epilepsy. The AHT’s method of mapping does not rely on collecting and amassing family samples and data. When an epileptic is identified it is not necessary to sample siblings or parents, nor does this particularly help. The method relies on excluding all differences except the one (or several) differences that make that dog epileptic. Epileptic identification is very important i.e. has the dog been diagnosed through a logical sequence of diagnostic tests. Epilepsy is a diagnosis of exclusion so all other causes of seizures must be ruled out before the dog is found to be epileptic.
It is true that a gene screen has failed once but time has moved on & methods are improving all the time. We have the money now to do this again but we don’t have very much time. So please, if you are reading this and have an affected keesie, please send in a sample.
Keeshond Breed Health Coordinator to The Kennel Club