Leslie A. Lyons, PhD
Department of Population Health & Reproduction, School of Veterinary Medicine, University of California - Davis, Davis, CA, USA
Review cat sequencing projects
Introduce concept of linkage disequilibrium
Review stratification of cat populations and breeds
Examine the power of case - control studies and the cat array
For the cat, two NIH-funded and a project supported by Hill's Pet Food, Inc. have produced the genome sequence from one inbred Abyssinian cat ("Cinnamon"). Overall, a ~14x genome coverage has been established for the cat. The deep coverage supported selection of single nucleotide polymorphism (SNPs) that are valid and the deep coverage helps the identification of SNPs that are in more than one breed to support the selection of SNPs with high minor allele frequencies (MAF). A resource called a DNA array or DNA chip can then be produced that contains assays for the highly polymorphic and evenly dispersed SNPs; thus these arrays can assess the entire genome of the cat in one experiment. Hill's Pet Nutrition, Inc. has also provided funding to support feline genome resource development;12 a cat SNP-based DNA array was commissioned and released in early 2012 from Illumina, Inc. The major benefit of the arrays is that they allow assessment of the entire genome: genome-wide association studies (GWASs). Because the SNPs are at such a high density, the cats used for a GWAS can be from a population, not direct relatives. Thus, individual cases of diseases or traits can be examined from a population or across breeds and populations; cases (cats with the trait) and controls (cats without the trait) are required. The deeper sequencing of the cat genome and the investigation of variation by resequencing in different cat breeds have allowed a great deal of progress in feline genetics, from the analysis of single gene traits to the investigation of more complex traits. The available genetic resources for the cat now are no longer a research bottleneck for feline studies; however, the acquisition of appropriate patients for sufficient cases and controls remains a rate limiting step. Hence the primary care veterinarian, veterinary specialists, and veterinary researchers need to join forces to properly characterize diseases and routinely collect research materials so that patients are not lost to important studies and health investigations.
Cat DNA Sequencing Projects
Case-control and affected pedigree member (APM) genome-wide association (GWA) studies are well established genetic approaches that have recently become available for companion animals1–9. The extended linkage disequilibrium found in companion animal breeds greatly increases the efficiency of GWA studies. The genome sequencing projects of companion animals have facilitated the resources needed for GWA style research. The dog has a 7.5x genome coverage that discovered 2.5 million SNPs across eleven breeds and wild canids (coyote and wolf).10 The horse has a 6.8x genome coverage that discovered 1 million SNPs across 7 breeds, one individual per breed.11 For the cat, two NIH-funded and a project supported by Hill's Pet Food, Inc. have produced the genome sequence from one inbred Abyssinian cat ("Cinnamon").12,13 Six different breeds were used for SNP discovery, four cats pooled in a breed. The first NIH project was led by AgenCourt and the Broad Institute at MIT using a Sanger-based approach.12 The newer NIH funded effort is using newer sequencing techniques, such as Solexa (Illumina) and Roche 454, collectively termed NextGen sequencing. Hill's Pet Food, Inc. added additional Sanger-based sequencing from a random bred cat, a South African wildcat, and one cat from 5 breeds.13 Overall, a ~14x genome coverage has been established for the cat, which is in the final stages of assembly by Washington University at St. Louis.
The deep coverage supports selection of single nucleotide polymorphism (SNPs) that are valid and the deep coverage helps the identification of SNPs that are in more than one breed to support the selection of SNPs with high minor allele frequencies (MAF). In addition, the 2x Sanger-based framework12 has assisted the assembly of the shorter reads of the NextGen technologies, facilitating the selection of SNPs with even spacing across the genome. The cat had 9.5 million SNPs overall for selection, which was 9-times better than the horse and ~4-times better than the dog. Thus, the cat array should be a significantly more powerful than the similar sized arrays of the horse and dog.
The Lyons' Feline Genetics laboratory at UC Davis has been preparing for GWA studies in the cat for several years. The laboratory has an inventory of over 15,000 phenotyped cat samples and access to an additional 15,000 via the UC Davis Veterinary Genetics Laboratory, which is one of the largest commercial genetics service centers for cats. The large sample inventory will facilitate genetic test development if GWA studies are successful. The laboratory has had NIH funding for cat resource development and disease model development and participated in the MAF feline advisory committee for the cat array development. The laboratory provided two of seven populations of cats that were part of the NIH-NHGRI funded cat sequencing effort. In addition, the Lyons lab provided 349 of the 384 samples required by Illumina for the GenTrain of the cat DNA array. These data can be potentially used for controls across several studies and ensure the laboratory has the first access to the nuances of the SNP characteristics and clustering patterns for the cat Illumina array.
Besides the power of the arrays, the extent of LD is important to the study design for GWA studies. In dog, LD ranges from 100 Kb in Labradors to over 1 to 4 Mb in many breeds.10 In comparison, humans have LD < 100 Kb. LD has also been estimated for the horse, with Quarterhorse having the weakest LD and Thoroughbreds having the strongest, horses having an average LD of twice the background level up to 100 to 150 Kb.11 One of the Lyons' laboratory research priorities has been an estimation of the LD across cat breeds. SNPs (N = 1536) were selected in 10 different 1 Mb regions in the cat genome, thus approximately 150 SNPs per regions. Twenty-two cat breeds were examined, including two breeds that had both USA and non-USA representatives to allow a USA vs. non-USA estimation of LD and two random bred populations (Eastern and Western) to obtain a base-line estimate of cat LD. An LD estimate was determined for 20 Burmese from the UK and 20 Burmese from the USA. As compared to horses11 and dogs (not shown), cat LD is not as strong (Average R2) below 100 Kb and all three species have a significant decrease after ~100 Kb. Burmese from the USA, which interestingly is one of the breeds with the least genetic diversity, have the strongest LD in the cat of the 22 breeds examined. Non-USA Burmese have average LD for cats, a bit stronger than the average horse breed after 100 Kb. These data suggest cat GWA studies will need more cases and controls than dog studies.
To date, over a dozen GWA studies have been published in dogs, mainly using the Affymetrics and Illumina arrays.14–20 These studies have primarily focused on single gene traits that have had relatively straightforward diagnoses. The regional localizations of the GWA studies have been between 760 Kb to 19.5 Mb for the dog and a broad association defined in the published horse study.21 The studies with more animals are more likely to have a more narrow localization, which greatly facilitates candidate gene screening. Once a GWA is performed, a significant amount of research may still be required to identify causative mutations for the disease in question, especially if the associated region is large.
Other research has focused on the population genetics of the cat that will facilitate the GWA studies by determining population stratification. An in-depth population study of 29 breeds (N = 477) and 37 worldwide cat populations (N = 944) has been performed by analyzing their genetic diversity and population structuring using STRs, SNPs, and mtDNA control region.22,23 This study clearly indicates a strong and extremely significant stratification between Western and Eastern (South Asian) cat populations and their respective breeds. Some breeds have clearly influenced the development of others to such a great extent that they cannot be genetically differentiated, such as Persians, Exotic Shorthairs, Scottish Folds, and British Shorthairs. Similarly, several Asian-derived breeds cannot be differentiated from the Siamese breed.
Cat DNA Array
An important by-product of the DNA sequencing effort is the identification of the normal genetic variation in the cat genome, SNPs. The SNPs can be verified to be specific to one breed or common across many breeds and populations. The genome assembly supports the proper positioning of the SNPs across the genome. A resource called a DNA array or DNA chip can then be produced that contains assays for the highly polymorphic and evenly dispersed SNPs; thus these arrays can assess the entire genome of the cat in one experiment. Hill's Pet Nutrition, Inc. has also provided funding to support feline genome resource development;12 a cat SNP-based DNA array was commissioned and released in early 2012 from Illumina, Inc.Each DNA chip, which is about the size of a microscope slide, has 12 regions; each region is used to test one cat. Each region has the assays for approximately 63,000 SNPs. The major benefit of the arrays is that they allow assessment of the entire genome; genome-wide association studies (GWASs). Because the SNPs are at such a high density, the cats used for a GWAS can be from a population, not direct relatives. Thus, individual cases of diseases or traits can be examined from a population or across breeds and populations; cases (cats with the trait) and controls (cats without the trait) are required. In addition, because there is less concern for the mode of inheritance of the trait, a GWAS can be performed even with traits that may have complex inheritance but a high heritability or relative risk in a population. Fewer cases are required to investigate a recessive trait, more for a dominant trait, and even more for complex traits that cause an increased relative risk - the more cases, the lower the relative risk.
Future of Cat Genetics
The deeper sequencing of the cat genome and the investigation of variation by resequencing in different cat breeds have allowed a great deal of progress in feline genetics, from the analysis of single gene traits to the investigation of more complex traits. However, many of the common diseases that plague humans are also found in cats will likely be examined in the outbred populations of non-pedigreed housecats because only 10% to 15% of cats in the United States are representatives of a fancy breed, a proportion that is higher than most other nations.21 Our random-bred/alley/moggy housecats are sharing our sedentary and indoor lifestyle as well as the associated health problems, such as diabetes, obesity, and asthma. Cats are obligate carnivores and require very high protein levels for normal nutrition. Increased fats and carbohydrates in pet foods lower the cost but can jeopardize the cat's health. Increases in the prevalence of feline inflammatory bowel disease and feline lower urinary tract disease are affected by commercial food qualities. Even though pet food companies do make enormous efforts to provide balanced nutrition for our companion animals, cats seem to be having complications with the transition from a wild-prey diet. Food allergies are of particular concern in cats, leading to the development of a wealth of alternative protein diets. Thus, genes involved in complex dietary interactions will be important in future studies.
Disease resistances and susceptibilities are also important to the future of feline genetics. Susceptibility to feline immunodeficiency virus (FIV) and particularly to disease caused by feline coronavirus is likely to be of particular interest. Although FIV has low morbidity and mortality rates in the cat, the genes influencing the cat's tolerance of FIV could shed light on interactions in humans and other species with similar immune-compromising pathogens. Feline enteric coronavirus is nearly ubiquitous in domestic cats. As an enteric pathogen, the virus may cause some malaise and diarrhea, but it is otherwise innocuous. However, mutated viral forms cause feline infectious peritonitis (FIP), which has a nearly 100% mortality rate in domestic cats, regardless of race, color, or breed. Deciphering the genes involved with infection and disease progression for FIP would be a major advancement for feline health. The cat genome sequence and the DNA arrays will greatly facilitate these studies.
Once causative mutations for heritable conditions are identified in the cat, cats become a more important asset to human health. Gene therapy approaches are already being explored for several inborn errors of metabolism in cats.11 Cats will become a more useful alternative and supportive animal model than rodent models for many heritable conditions for reasons such as the following:
Cats provide a balance between cost and efficiency.
Drug dosages are more easily translated between cats and humans.
The longer life span of the cat allows repeated therapy trials and longer term studies.
Cats have strong conservation of biology, anatomy, and physiology with humans.
Cats provide a second animal for validation and efficacy.
The larger size of the cat and its organs are more amenable to therapies.
Finally, cats are intermediate with regard to genetic variation, mimicking human populations and ethnic groups more closely than inbred strains of mice. For example, many murine models exist for the study of cystogenesis, the hallmark of PKD; however, each rodent model has its shortcomings. PKD in cats is similar to human autosomal dominant PKD in several important aspects, including the following: (1) a causative mutation in PKD1; (2) a similar type of mutation that causes a similar protein disruption; (3) similar variability in disease progression; (4) cystogenesis in other organs, including the liver and pancreas; and (5) the fact that homozygosity for the mutation is lethal.
The available genetic resources for the cat now are no longer a research bottleneck for feline studies; however, the acquisition of appropriate patients for sufficient cases and controls remains a rate limiting step. Hence the primary care veterinarian, veterinary specialists, and veterinary researchers need to join forces to properly characterize diseases and routinely collect research materials so that patients are not lost to important studies and health investigations. The development of the DNA tests for parentage and identification (VIN editor: the original link http://www.isag.org.uk/ has moved to http://www.isag.us/index.asp), coat colors, and the prominent diseases (e.g., PKD and hypertrophic cardiomyopathy) has encouraged cat breeders to explore genetic research more openly and has encouraged their participation in research studies. For these reasons more cat breeders are banking DNA material from their animals and providing DNA to service and research laboratories. Many veterinary hospitals and large clinical conglomerates are developing electronic database systems that could facilitate the identification of proper patients, cases, and controls. Combined with DNA banking and specialty health care, the veterinary world stands to enhance the possibilities of complex disease research in the cat by leaps and bounds. Even though the origins of the cat remain a mystery and domesticated may not be the most appropriate term for the domestic cat, researchers are unlocking its genetic secrets to explain its form and function.
1. Weeks DE, Lange K. The affected-pedigree-member method of linkage analysis. Am J Hum Genet 1988;42:315-326.
2. Teng J, Risch N. The relative power of family-based and case-control designs for linkage disequilibrium studies of complex human diseases. II. Individual genotyping. Genome Res 1999;9:234–241.
3. Risch NJ, Zhang H. Mapping quantitative trait loci with extreme discordant sib pairs: sampling considerations. Am J Hum Genet 1996;58:836–843.
4. Risch N, Zhang H. Extreme discordant sib pairs for mapping quantitative trait loci in humans. Science 1995;268:1584–1589.
5. Risch N. Linkage strategies for genetically complex traits. III. The effect of marker polymorphism on analysis of affected relative pairs. Am J Hum Genet 1990;46:242–253.
6. Risch N. Linkage strategies for genetically complex traits. II. The power of affected relative pairs. Am J Hum Genet 1990;46:229–241.
7. Risch N. Linkage strategies for genetically complex traits. I. Multilocus models. Am J Hum Genet 1990;46:222–228.
8. Risch N. Corrections to "Linkage strategies for genetically complex traits. III. The effect of marker polymorphism on analysis of affected relative pairs". Am J Hum Genet 1992;51:673–675.
9. Hauser ER, Boehnke M, Guo SW, et al. Affected-sib-pair interval mapping and exclusion for complex genetic traits: sampling considerations. Genet Epidemiol 1996;13:117–137.
10. Lindblad-Toh K, Wade CM, Mikkelsen TS, et al. Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature 2005;438:803–819.
11. Wade CM, Giulotto E, Sigurdsson S, et al. Genome sequence, comparative analysis, and population genetics of the domestic horse. Science 2009;326:865–867.
12. Pontius JU, Mullikin JC, Smith DR, et al. Initial sequence and comparative analysis of the cat genome. Genome Res 2007;17:1675–1689.
13. Mullikin JC, Hansen NF, Shen L, et al. Light whole genome sequence for SNP discovery across domestic cat breeds. BMC Genomics 2010;11:406.
14. Wilbe M, Jokinen P, Truve K, et al. Genome-wide association mapping identifies multiple loci for a canine SLE-related disease complex. Nat Genet 2010;42:250–254.
15. Vage J, Wade C, Biagi T, et al. Association of dopamine- and serotonin-related genes with canine aggression. Genes Brain Behav 2010;9:372–378.
16. Wiik AC, Thoresen SI, Wade C, et al. A population study of a mutation allele associated with cone-rod dystrophy in the standard wire-haired dachshund. Anim Genet 2009;40:572–574.
17. Awano T, Johnson GS, Wade CM, et al. Genome-wide association analysis reveals a SOD1 mutation in canine degenerative myelopathy that resembles amyotrophic lateral sclerosis. Proc Natl Acad Sci USA 2009;106:2794–2799.
18. Wiik AC, Wade C, Biagi T, et al. A deletion in nephronophthisis 4 (NPHP4) is associated with recessive cone-rod dystrophy in standard wire-haired dachshund. Genome Res 2008;18:1415–1421.
19. Karlsson EK, Baranowska I, Wade CM, et al. Efficient mapping of Mendelian traits in dogs through genome-wide association. Nat Genet 2007;39:1321–1328.
20. Bannasch D, Young A, Myers J, et al. Localization of canine brachycephaly using an across breed mapping approach. PLoS One 2010;5(3):e9632.
21. Lykkjen S, Dolvik NI, McCue ME, et al. Genome-wide association analysis of osteochondrosis of the tibiotarsal joint in Norwegian Standardbred trotters. Anim Genet 2010;41(Suppl 2):111–120.
22. Lipinski MJ, Froenicke L, Baysac KC, et al. The ascent of cat breeds: genetic evaluations of breeds and worldwide random-bred populations. Genomics 2008;91:12–21.
23. Grahn RA, Kurushima JD, Billings NC, et al. Feline non-repetitive mitochondrial DNA control region database for forensic evidence. Forensic Sci Int Genet 2010;5(1):33–42.