The Genetics of Gastric Dilatation and Volvulus (Bloat) in Dogs: What Do We Know and Where Are We Going?
Claire R. Sharp, BSc, BVMS (Hons), MS, DACVECC
School of Veterinary and Life Sciences, Murdoch University, Perth, WA, Australia; Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA, USA
The goal of this lecture is to review some of what we know about the potential genetic contribution to gastric dilatation and volvulus (GDV), commonly called bloat, in dogs. In addition, we will discuss funding by the AKC Canine Health Foundation (CHF) Bloat Initiative that aims to advance our understanding of the causes of bloat and how this funding is being utilized.1
Gastric dilatation and volvulus is a common condition in large and giant breed dogs with an unacceptably high morbidity and mortality rate. Based on the current veterinary literature published in peer-reviewed journals, the breeds in which GDV is documented most commonly include German Shepherd dog, Great Dane, Standard Poodle, Labrador Retriever, Akita, Golden Retriever, Saint Bernard, Doberman Pinscher, ChowChow, Collie, Rottweiler, Mastiff, Weimaraner, Bloodhound, Basset Hound, Belgian Shepherd, Great Pyrenees, Boxer, Husky, German Shorthaired Pointer, Samoyed, Newfoundland, and Bernese Mountain Dog.2 Other purebred dogs in which GDV has been reported include the Airedale Terrier, Anatolian Shepherd, Bernese Mountain Dog, Borzoi, Bouvier des Flandres, Briard, Bullmastiff, Cane Corso, Chesapeake Bay Retriever, Chinese Shar-Pei, Curly Coated Retriever, Dogue de Bordeaux, Flat-coated Retriever, Giant Schnauzer, Gordon Setter, Greater Swiss Mountain Dog, Irish Red and White Setter, Irish Setter, Irish Wolfhound, Komondor, Leonberger, Old English Sheepdog, Otterhound, Scottish Deerhound, Spinone Italiano, and Sussex Spaniel.2 The clear breed predisposition to GDV is the first suggestion that there may be a genetic component to GDV causation. Given the potential for a genetic contribution to the disease and since this is a disease of significance to so many dog breeds, there has been interest in genetic studies of bloat for decades. Unfortunately, genetic testing has not been adequately advanced until recently to be able to investigate complex genetic diseases. Fortunately, we are now in an era where genetic analyses are adequately advanced, available, and affordable to allow researchers to start to get to the bottom of the genetics of bloat.
If GDV does have a genetic contribution, understanding the genetics would ultimately allow us to institute selective breeding to reduce the disease prevalence. With GDV, as with many diseases, prevention is better than cure. Retrospective evaluation of 498 dogs with GDV seen at the Tufts veterinary school over a 10-year period documented that 23.5% are euthanized at presentation to the hospital; fewer die (6.0%) or are euthanized during the course of treatment (7.8%), resulting in an overall mortality rate of 37.3% (unpublished data). These data suggest that focusing research efforts on disease prevention will benefit a much greater number of dogs than focusing research on treatment strategies after the disease has occurred. Unfortunately, our current knowledge of the pathogenesis of GDV is inadequate to allow us to develop comprehensive strategies for disease prevention.
Due to the importance of GDV in many dog breeds, several large previous studies have investigated risk factors for the development of GDV in affected breeds.1-7 It is known that there is no single cause, genetic or otherwise, for GDV; rather, its occurrence is multifactorial. Although different studies have slightly different findings, generally considered risk factors for developing GDV include first-degree relatives that have had GDV, higher thoracic depth-to-width ratio, lean body condition, advancing age, eating quickly, fearful, nervous or aggressive temperament; and several diet-related factors, including being fed only dry food and/or a single large meal each day.1-6 These risk factors suggest that GDV results from both genetic and environmental factors. A genetic contribution should explain why the disease runs in breeds and runs in families. Genetics may also explain part of the thoracic depth:width ratio and even contribute to temperament that in turn affects GDV risk. It is now time to further investigate GDV pathogenesis through the application of genomic and molecular methods to understanding the basis for these risks.
It is known that GDV is not a single gene disorder, fitting a simple pattern of Mendelian inheritance. However, given that GDV "runs in families" of dogs, it is very likely to be a complex, multifactorial (or polygenic) disorder; that is, it is likely associated with the effects of multiple genes in combination with environmental factors and 'lifestyle.' Current recommendations are that, as with other polygenic disorders, the breadth of the pedigree increases selective pressure against the condition. Breeders are encouraged to select dogs for breeding with lower thoracic depth-width ratios and whose littermates have not had GDV;8 however, given the potentially delayed age of onset of GDV, this is challenging. Unfortunately, rather than the prevalence of GDV decreasing with the knowledge of the aforementioned predispositions and the potential for adjusted breeding practices, it actually appears to be increasing. Additionally, despite the well-known fact that predisposition to GDV is heritable, even within at-risk breeds, and the increasing prevalence of disease despite breeding recommendations, no genetic studies of GDV have been published in the peer-reviewed literature to date. It is clear that further investigation is needed into the cause of GDV so that more clear guidelines can be provided to breeders and the disease can be prevented in much greater numbers of dogs.
Some studies appear to have been conducted and published online; however, limited data are available for rigorous scrutiny in peer-reviewed journals. In the late 1990s, Dr. John Armstrong, a geneticist with a particular interest in dog genetics and bloat, interrogated some Standard Poodle pedigrees for bloat.9 An example of one pedigree analysis is shown below.9
Pedigree analysis in this form is most useful for traits that follow classic Mendelian inheritance, rather than for complex polygenic disorders as GDV is expected to be. In this pedigree an argument could be created that the trait is either dominant with incomplete penetrance or even recessive (if all unaffected dogs were heterozygous carriers). Unfortunately, these pedigrees don't really help us get to the bottom of the disease. Such analyses were performed prior to the dog genome project that now allows us to be able to perform genome-wide association studies (GWAS). Genome-wide association studies will form the basis of some of the ongoing research into GDV genetics.
Similar pedigrees have been analyzed for Irish Setters, also in the late 1990s.5 One study showed that within Irish Setters having a relative with GDV (particularly a parent) increased the GDV risk.5 Additionally, 5 generation pedigrees yielded a significantly higher mean coefficient of relationship for 11 dogs with GDV than 11 dogs without GDV.5
Beyond pedigree analysis, investigation of the genetic basis of bloat could take the form of either candidate gene studies or GWAS. Although it seems unlikely from epidemiologic studies that GDV is a single gene disorder, candidate gene studies (i.e., those investigating the role and function of single genes) are being undertaken and may provide some useful insight. Although not yet published, a progress report on a study being performed by Dr. Michael A. Harkey at the Fred Hutchinson Cancer Research Center in Seattle, WA was recently released (on 7/31/15).10 In this report, provided to Great Dane owners involved in the study, Dr. Harkey provided some information about candidate gene studies investigating MHC genes in Great Danes affected by GDV (vs. control dogs).10 Dr. Harkey's hypothesis is that GDV is caused by an imbalance in the bacterial population of the gut (i.e., the gut microbiome), and that this imbalance is caused by a specific array of genes encoding the major histocompatibility complex (MHC).10 MHC is an important part of a dog's immune system that is involved in our bodies' ability to differentiate self from non-self (non-self being things like bacterial invaders). Mutations in MHC genes have also been implicated in other diseases in dogs, such as Addison's disease, systemic lupus erythematosus (SLE), immune-mediated meningitides, immune-mediated hemolytic anemia, and anal furunculosis amongst others. Dr. Harkey's research is specifically focused on two of the MHC genes, DLA88 (dog leukocyte antigen 88) and DRB1.10 With 75 dogs, his study has already identified evidence that these two MHC genes play a role in predisposing dogs to bloat.10 For example, this research has identified a new variant of the DLA88 gene that has not been reported in dogs before; and this variant (FB) is three times more prevalent in Great Danes with bloat than control dogs.10 Another variant in DLA88 (1001) is 7 times more common in unaffected Great Danes than affected dogs, and thus is thought to be protective against bloat.10 Additionally, a variant in DRB1 (1201) is seen twice as often in GDV-affected Danes than controls.10 This research group is continuing to add additional cases to their studies (including other breeds) and so we eagerly await the publication of these initial results and ongoing study findings.
Another investigator performing candidate gene studies is Dr. Laura Nelson. Dr. Nelson received funding from the AKC CHF Bloat Initiative commencing in 2014 for her study entitled "Abnormalities in the stomach's ability to contract predisposes large-breed dogs to bloat."11 This study has two components, the first being to assess stomach and gastrointestinal motility in dogs predisposed to bloat using the noninvasive methodology SmartPill®. In addition, Dr. Nelson will investigate the potential for genetic abnormalities in the motilin gene to predispose dogs to bloat.11 Motilin is a hormone that contributes to stomach motility. Abnormalities in motilin secondary to changes in the motilin gene have been identified in cattle with a similar condition to GDV called LDA (left displaced abomasum).12 Additionally, other single nucleotide polymorphisms have been associated with bovine abomasal displacement that may provide useful candidate genes for further studies in dogs.12
Rather than candidate gene studies, our group is focusing on GWAS as a mechanism to explore GDV causation. Genome-wide association studies aim to look for genomic signatures associated with disease anywhere in the genome. My research group comprising Drs. Jerry Bell, Liz Rozanski (also at Tufts), Kerstin Lindblad-Toh (from the Broad Institute) and Steve Hannah (from Purina) has undertaken a study, also as part of the bloat initiative entitled "Evaluating the complex genetic basis of bloat."11 One of the goals of our study is to identify (through a GWAS) single nucleotide polymorphisms (SNPs) that are more common in dogs that develop GDV than those that do not. Additionally, epigenetic phenomena (i.e., functionally relevant modifications in the genome that do not involve a change in the nucleotide sequence) are also worthy of investigation, and it is expected that epigenetic mechanisms are likely to explain some of the environmental, age- and diet-related influences on GDV risk.
We know from extensive investigation of complex diseases in human medicine that GWAS provide information about disease-associated loci but give only weak mechanistic insight. Thus, the best approach to understanding disease causality is a systems biology approach. Systems biology strategies integrate large-scale GWAS that define the genotype and functional genomics that define the intermediate molecular phenotypes, with clinical disease phenotypes to elucidate disease pathogenesis.9 Since heritable information flows from DNA to mRNA to proteins and then to phenotypic traits, evaluating each aspect of this system is required for a holistic understanding of disease. The overriding goal of this study is to take a systems biology approach to understanding the pathogenesis of GDV in dogs.
We hypothesize that GDV is a multilayer complex disease whose causality will be explained by changes in the genome, epigenome, transcriptome, proteome and metabolome, and that the mechanism will be most clearly described when all data types are integrated in a large-scale systems biology approach. Specifically, we hypothesize that we will be able to explain a substantial proportion of disease risk with SNPs identified in a GWAS and an additional proportion of disease risk by identifying intermediate phenotypes that collectively represent the disease association downstream of genomic DNA such as the epigenome, transcriptome, proteome and metabolome. With this integrative or systems biology approach we hope to be able to ultimately pinpoint the key regulatory genes and mechanisms of disease pathophysiology.
Given the expected significant genomic differences between breeds, we propose to conduct the discovery phase of this first genomic study of GDV in a single breed of dog. Review of 498 cases of GDV over the last 10 years at our institution revealed that by far the most represented breed is the German shepherd dog (GSD), accounting for 20.9% of cases at our institution. Thus, the GSDs enrolled will be used for the discovery cohort in this study. Subsequently, validation of our findings will be performed in other purebred dogs affected by GDV.
The specific aims of the study are:
1. To establish a biobank (a repository of biological samples with associated data) for use in this and future GDV research.
2. To identify susceptibility genes for GDV through a discovery genome-wide association study in German shepherd dogs and a validation study in other purebred dogs with GDV.
3. To perform gene expression profiling (transcriptomics) in blood and target cells/tissues (PBMCs, gastric smooth muscle, and gastric ligaments), and compare gene expression between dogs with GDV and control dogs.
4. To identify epigenetic modifications associated with differential gene expression in GDV, specifically evaluating DNA methylation and histone modification.
5. To perform proteomics using Millipore multiplex assays for evaluation of key cytokines and gastrointestinal hormones for comparison between dogs with GDV and control dogs.
6. To characterize the metabolome in dogs with GDV.
Although no data/results from the Bloat Initiative Grants are yet available, we have received overwhelming support from pet owners, dog breeders, breed clubs, and veterinarians from around the country, and have enrolled more than 200 dogs to date. We are hopeful that as we gather greater numbers and start to analyze genetic information we will start to elucidate some of the genetics that underlie GDV risk. Our ultimate goal is to develop genetic tests for genes that increase GDV risk within affected breeds, so that breeders are able to test their breeding dogs and employ selective breeding to reduce the prevalence of undesirable genes in their bloodlines. Additionally, having genetic tests available will allow us to identify which dogs would benefit most from preventative procedures such as prophylactic gastropexy.
References
1. Glickman LT, Glickman NW, Perez CM, Schellenberg DB, Lantz GC. Analysis of risk factors for gastric dilatation and dilatation-volvulus in dogs. J Am Vet Med Assoc. 1994;204(9):1465–1471.
2. Bell JS. Inherited and predisposing factors in the development of gastric dilatations and volvulus. Top Companion Anim Med. 2014;29(3):60–63.
3. Glickman LT, Glickman NW, Schellenberg DB, Raghavan M, Lee TL. Incidence of and breed-related risk factors for gastric dilatation-volvulus in dogs. J Am Vet Med Assoc. 2000;216(1):40–45.
4. Glickman LT, Glickman NW, Schellenberg DB, Raghavan M, Lee T. Non-dietary risk factors for gastric dilatation-volvulus in large and giant breed dogs. J Am Vet Med Assoc. 2000;217(10):1492–1499.
5. Schellenberg DB, Yi Q, Glickman NW, Glickman LT. Influence of thoracic conformation and genetics on the risk of gastric dilatation-volvulus in Irish Setters. J Am Anim Hosp Assoc. 1998;34(1):64–73.
6. Raghavan M, Glickman NW, Glickman LT. The effect of ingredients in dry dog foods on the risk of gastric dilatation-volvulus in dogs. J Am Anim Hosp Assoc. 2006;42(1):28–36.
7. Pipan M, Brown DC, Battaglia CL, Otto CM. An Internet-based survey of risk factors for surgical gastric dilatation-volvulus in dogs. J Am Vet Med Assoc. 2012;240(12):1456–1462.
8. Bell JS. Risk factors for canine bloat. In: Proceedings from the Tufts' Canine and Feline Breeding and Genetics Conference; 2003. Tufts Cummings School of Veterinary Medicine.
9. Armstrong J. Pedigree analysis: bloat in the Standard Poodle. The Institute of Canine Biology. Accessed: 20 July 2015. www.instituteofcaninebiology.org/bloat-in-the-standard-poodle.html.
10. Harkey MA. Update on the canine bloat study 7/31/15. Fred Hutch Cancer Research Center.
11. Bloat Initiative Grants. American Kennel Club Canine Health Foundation. Accessed 20 July 2015. www.akcchf.org/canine-health/your-dogs-health/bloat/bloat-initiative-grants.html.
12. Zebrin I, Lehner S, Distl O. Genetics of bovine abomasal displacement. Vet J. 2015;204:17–22.