(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Genetic epidemiology of blood type, disease and trait variants, and genome-wide genetic diversity in over 11,000 domestic cats [1]

['Heidi Anderson', 'Wisdom Panel Research Team', 'Wisdom Panel', 'Kinship', 'Portland', 'Oregon', 'United States Of America', 'Stephen Davison', 'Katherine M. Lytle', 'Leena Honkanen']

Date: 2022-08 

In the largest DNA-based study of domestic cats to date, 11,036 individuals (10,419 pedigreed cats and 617 non-pedigreed cats) were genotyped via commercial panel testing elucidating the distribution and frequency of known disease, blood type, and physical trait associated genetic variants across cat breeds. This study provides allele frequencies for many disease-associated variants for the first time and provides updates on previously reported information with evidence suggesting that DNA testing has been effectively used to reduce disease associated variants within certain pedigreed cat populations over time. We identified 13 disease-associated variants in 47 breeds or breed types in which the variant had not previously been documented, highlighting the relevance of comprehensive genetic screening across breeds. Three disease-associated variants were discovered in non-pedigreed cats only. To investigate the causality of nine disease-associated variants in cats of different breed backgrounds our veterinarians conducted owner interviews, reviewed clinical records, and invited cats to have follow-up clinical examinations. Additionally, genetic variants determining blood types A, B and AB, which are relevant clinically and in cat breeding, were genotyped. Appearance-associated genetic variation in all cats is also discussed. Lastly, genome-wide SNP heterozygosity levels were calculated to obtain a comparable measure of the genetic diversity in different cat breeds. This study represents the first comprehensive exploration of informative Mendelian variants in felines by screening over 10,000 pedigreed cats. The results qualitatively contribute to the understanding of feline variant heritage and genetic diversity and demonstrate the clinical utility and importance of such information in supporting breeding programs and the research community. The work also highlights the crucial commitment of pedigreed cat breeders and registries in supporting the establishment of large genomic databases, that when combined with phenotype information can advance scientific understanding and provide insights that can be applied to improve the health and welfare of cats.

Domestic cats are one of the world’s most popular companion animals, of which pedigreed cats represent small unique subpopulations. Genetic research on pedigreed cats has facilitated discoveries of heritable conditions resulting in the availability of DNA testing for studying and managing inherited disorders and traits in specific cat breeds. We have explored an extensive study cohort of 11,036 domestic cat samples representing pedigreed cats of 90 breeds and breed types. This work provided insight into the heritage of feline disease and trait alleles. We gained knowledge on the most common and relevant genetic markers for inherited disorders and physical traits, and the genetic determinants of the clinically relevant AB blood group system. We also used a measure of genetic diversity to compare inbreeding levels within and between breeds. This information can help support sustainable breeding goals within the cat fancy. Direct-to-consumer genetic tests help to raise awareness of various inherited single gene conditions in cats and provide information that owners can share with their veterinarians. In due course, ventures of this type will enable the genetics of common complex feline disease to be deciphered, paving the way for precision healthcare with the potential to ultimately improve welfare for all cats.

Our previous work has elucidated that the comprehensive screening of genetic variants in dogs is convenient and justified as it provides information to support breeding programs, veterinary care and health research [ 31 ]. Further large-scale multiplex screening approaches were taken to characterize canine disease heritage and the relative frequency and distribution of disease-associated variants across breeds and to explore the frequency of known canine appearance-associated variants among dog breeds [ 32 , 33 ]. As we will demonstrate, such efforts are now equally feasible in cats and hold comparable promise for gaining insight into the genetic epidemiology of feline diseases and traits to better inform feline breeding decisions and establish the foundation for precision medicine of individuals, populations, and breeds.

Domestication of cats occurred approximately 10,000 years ago [ 10 ], with pedigreed cats emerging over the last 150 years representing genetically unique domestic cat subpopulations. The majority of the approximately 100 cat breeds, breed types (or varieties of the breed) known today, including several new breeds being developed, represent breeds that are less than 75 years old. The genetic research of isolated populations in humans, dogs, and more recently cats has vastly improved identification of genetic disease and trait variants for use in genetic diagnostics and medicine. As an example, through studies of pedigreed cats, researchers have identified disease-associated variants for both diseases that are affecting cats worldwide such as Polycystic Kidney Disease and rare disorders within a specific breed such as Hypotrichosis in Birman cats [ 11 – 19 ]. Moreover, studies of cat breeds have enabled discovery of the genetic variants determining blood types of the AB blood group system that associate with neonatal isoerythrolysis and hemolytic transfusion reactions; and many of the physical traits, which are an integral part of breed development [ 8 , 20 – 25 ]. Research on genetically isolated populations has further highlighted that reduced genetic diversity, because of isolation and generations of selective breeding to consistently produce animals with uniform appearance, can manifest as an increase in the number of health conditions that are identified [ 26 – 29 ]. For more than a decade, it has been common practice to eradicate disease-associated variants from pedigreed cat breeding populations using DNA testing. However, the focus on eradicating single DNA variants from a breed could contribute to severe loss of genetic diversity, especially if implemented strictly instead of thoughtfully [ 30 ]. Improved genomic perspective on domestic cats, and the breeds’ level of genetic diversity, have propelled a growing number of breeders to incorporate breeding strategies for sustaining genetic diversity and increasing it through crossbreeding. Crossbreeding in cat breeding is also used to introduce a trait that is new to the breed. In addition, the cat breeds that are becoming increasingly more common have a hybrid origin such as the world’s most popular cat breed, the Bengal, that is a result of breeding an Asian Leopard Cat with the domestic cat.

Domestic cats are valued companions to humans, with owners seeking high quality veterinary care to support long-term wellness and positive outcomes for disease treatment. Cat genomes are structured in a similar way to humans, with 90% of their genes having a human homologue, and cats and humans also suffer from many similar diseases [ 1 ]. Domestic cats, like dogs, can serve as naturally occurring models for many human diseases, in which the development of therapeutic treatments may be helpful for both veterinary and human patients [ 1 – 6 ]. Genomic medicine, encompassing the understanding of individual genetic variability in disease risk and the customization of healthcare by patient subgroups, is now feasible and likely to become a future standard-of-care in veterinary medicine and the care of companion animals, including cats [ 1 ]. Fueling this development for felines specifically are improved genomic resources and technologies, combined with a steady increase in the discovery of Mendelian disease and trait variants [ 7 , 8 ]. Genetic testing is a form of genomic medicine when used for diagnosing diseases or traits of clinical relevance, as the results are patient-specific and can potentially be used to tailor treatment to the disease and the patient [ 9 ]. Direct-to-consumer genetic testing is now readily available, and further empowers owners to proactively invest in potentially changing the medical care for their pet.

Medical history information for genetically affected cats was collected through interviews with cat owners and veterinary clinicians. Clinical examinations were performed for confirmation of a Factor XII deficiency diagnosis by collecting a blood sample for an activated partial thromboplastin time screening test through IDEXX Laboratories (IDEXX Europe B.V., Hoofddorp, The Netherlands). Progressive Retinal Atrophy diagnosis was confirmed via ophthalmic examination performed by an ECVO (European College of Veterinary Ophthalmologists) board-certified veterinary specialist. A set of blood type determination data was available through the records of the former Genlab Niini (Helsinki, Finland). All additional phenotype information (clinical or trait) and documentation was obtained through voluntary owner submissions.

Microarray genotyping analyses were carried out following manufacturer-recommended standard protocols for the Illumina XT platform (Illumina, Inc.). All genotype data from samples with call rates below 98% of the analyzed markers were discarded to ensure high quality data and all disease-associated variant calls were confirmed by manual review. Only disease and trait variants reaching 100% sensitivity and specificity upon testing with natural control samples that are homozygous/heterozygous for the disease or trait associated variant (70/87 variants), or validated using synthetic control oligonucleotides (17/87 variants), were considered for reporting. Genome-wide genetic diversity was measured as a percentage of heterozygous SNPs across a set of 7,815 informative SNPs. Data available at Dryad Digital Repository [ 34 ]. Selected disease-associated variant findings were genotyped using standard capillary sequencing on an ABI3730xl DNA Analyzer platform (Thermo Fisher Scientific, Waltham, MA, USA) as a secondary technology to provide further validation of results that were unexpected for the breed. Sequencing was performed at the Sanger Sequencing Unit of the Finnish Institute of Molecular Medicine (FIMM). The DNA extractions and PCR-product preparation and purification were carried out as previously described in detail [ 31 ] using ~20 ng of genomic template DNA and an Amplitaq Gold Master Mix-based protocol according to the manufacturer’s instructions (Applied Biosystems, Waltham, MA, USA).

A custom genotyping microarray for selected feline disease and trait associated variants ( S2 Table ) was developed based on the Illumina Infinium XT platform (Illumina, Inc., San Diego, CA, USA), commercially available as the Wisdom Panel Complete for Cats / MyCatDNA / Optimal Selection Feline tests. The microarray was designed and validated for use following the same protocol and principles as previously described for canines [ 21 ]. Firstly, public databases [ 8 ] and searches of the scientific literature were used to identify likely causal variants for feline Mendelian disorders and traits. Secondly, genotyping assays for the identified variants were designed according to the manufacturer’s guidelines (Illumina, Inc.). At least three technical replicates of each target sequence were included in the array design. Thirdly, the validation of each specific disease and trait assay was completed with the use of control samples of known genotype or synthetic oligonucleotides in the case of rare conditions for which no control samples were available. In addition, owner-provided photographs contributed to phenotypic validation of trait variant tests.

The breed of a cat was reported by its owner with accompanying registration information confirming the cat was registered with The International Cat Association (TICA), Fédération Internationale Féline (FiFe), The Cat Fanciers’ Association (CFA), or World Cat Federation (WCF) standards. Additional breeds not yet recognized by any major breed registry but with an established community of breed hobbyists were also considered breeds for the purposes of this study. The non-pedigreed cat sample set consisted of mixed breeds, breed crosses, or random-bred cats. The tested cats were most often from the United States of America (54.9%), while cats from Finland (17.4%), Canada (5.3%), United Kingdom (3.5%), Norway (3.5%) and Sweden (3.3%), Russia (2.5%) and France (1%) represented other notable subgroups (>1% of the sample).

The cat study population consisted of 10,419 pedigreed cats and 617 non-pedigreed cats whose samples were obtained during the development and provision of the commercially available MyCatDNA and Optimal Selection Feline tests (Wisdom Panel, Helsinki, Finland and Wisdom Panel, Vancouver, WA, USA, respectively) between 2016 and 2021. The 10,419 pedigreed cat samples represented 90 breeds and breed types (or variants of the breed) with 60 (66.6%) breeds and breed types represented by 15 or more individuals ( S1 Table ).

Feline DNA was obtained by Wisdom Panel as owner submitted, non-invasive cheek swab samples or was collected by certified veterinary clinics as cheek swab and blood samples in accordance with international standards for animal care and research. All cat owners provided consent for the use of their cat’s DNA sample in scientific research. University biobank samples were collected under the permit ESAVI/6054/04.10.03/2012 by the Animal Ethics Committee of the State Provincial Office of Southern Finland, Hämeenlinna, Finland.

Results

Overview of genotyping results A total of 11,036 domestic cats, mainly pedigreed cats (N = 10,419) representing 90 breeds and breed types (or variants of the breed) and an additional 617 non-pedigreed cats, were successfully genotyped on the custom array (the sample failure rate for DNA extracted from buccal swabs was <1%). Genetic screening included genotyping of 7,815 informative SNP markers across the genome and 87 variants associated with blood type, diseases, and/or physical appearance; 83 of which were evaluated in the entire 11,036 cat study sample and four more recently included variants to the genotyping platform that were screened for in a subset of 2,186 samples (Tables 1 and S1–S4). We observed 57 (65.5%) of the 87 tested variants at least once in this study cohort, including 22 (38.6%) disease-associated variants and 35 (61.4%) variants associated with an appearance-associated trait or blood type (Tables 1 and S2–S4). The maximum number of disease-associated variants observed in any one individual was four. We observed that 2,480 (22.5%) of the tested cats had at least one disease-associated variant present and 452 (4.1%) of the tested cats were potentially at risk for at least one health condition, in accordance with the disorders’ modes of inheritance. The maximum number of different disease-associated variants present in a single breed was 9; this was observed in the Maine Coon, which was the breed that was represented by the most individuals (N = 1971) tested in this dataset. Three genetic variants for the diseases Hyperoxaluria Type II [35], Lipoprotein Lipase Deficiency [36] and Myotonia Congenita [37], were solely observed in non-pedigreed cat samples. While we observed several disease-associated variants in the breeds with documented occurrence, we also detected 13 disease-associated variants in additional breeds or breed types in which the variants had not previously been reported. For each additional breed finding, extensive review of the breed information was performed, including use of the proprietary Wisdom Panel cat breed determination algorithm where necessary. Details of the variants found in additional breeds are listed in Table 2. Within these additional breeds, individuals that were genetically affected (having one copy of a dominant variant or two copies of a recessive variant) were identified and flagged for clinical follow-up. Individuals that were genetically affected for candidate disease-associated variants previously presented by a case study, were also shortlisted for follow-up. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 1. Prevalence and frequency of the tested disease, blood type and trait -associated variants in all cats. https://doi.org/10.1371/journal.pgen.1009804.t001 PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 2. Summary of disease-associated variant findings in additional breeds and breed types. https://doi.org/10.1371/journal.pgen.1009804.t002

Genetic epidemiology of the common AB blood group system across breeds and breed types The major feline AB blood groups including blood type A, blood type B and the rare blood type AB are caused by functional differences in the cytidine monophospho-N-acetylneuraminic acid hydroxylase enzyme, encoded by the CMAH gene, that impact the ability of the enzyme to convert sialic acid N-acetylneuraminic acid (Neu5Ac) to N-glycolylneuraminic acid (Neu5Gc) on erythrocytes [20–25]. For purpose-bred cats, the current convention (2019 typing panel recommendation) [23,24], proposes that genetic testing for blood types A, B and AB should be based on panel testing of the four following variants c.179G>T, c.268T>A, c.364C>T and c.1322delT of CMAH. In this study, we obtained genotypes for ten blood type-associated CMAH variants; c.1-53delGTCGAAGCCAACGAGCAA (18bp indel), c.139C>T, c.142G>A, c.179G>T, c.187A>G, c.268T>A, c.327A>C, c.364C>T, c. 1322delT and c.1603G>A. All 11,036 cats were genotyped for nine of these CMAH variants, while c.1322delT was more recently added to the genotyping platform providing genotype results for a subset of 2,186 cats (19.7%). To evaluate the suitability of the proposed ‘2019 typing panel’ as the blood type genotyping scheme, we compared the obtained genotype information to the immunological blood type results available for 220 cats of this study cohort (S5 Table). We found that blood types assigned according to the cat’s 2019 typing panel-based genotype showed a 99.1% concordance with immunologically determined blood types. One Ragdoll with the c/c genotype indicative of blood type AB [20,23,24], had been determined serologically to have blood type A according to the cat’s owner and one Donskoy with the A/b genotype indicative of blood type A had been determined serologically to have blood type AB. Retesting or further clinical investigation were not performed for these cats due to lack of availability. After determining the suitability of 2019 typing panel, we based genetic blood typing on the variants: c.268T>A, (b1); c.179G>T, (b2); c.364C>T, (c—resulting in blood type AB); and c.1322delT (b3). The variants b1, b2 and c were observed widely distributed across breeds with frequencies of 12.6%, 1.6% and 1.5% in all cats, respectively (S4 Table). No more than two variants in total were found in any individual cat. Though generally uncommon in all cats, variant b2 was a major variant associated with blood type in the breeds Chartreux, Donskoy, Egyptian Mau, Minuet Longhair, Pixiebob, Siberian, Tennessee Rex, Tonkinese, Toybob, Turkish Angora and Turkish Van. Variant b3 was exclusively found in Ragdolls (16.9% allele frequency) and in a Ragdoll mix across the subset of 2,186 genotyped cats representing 69 breeds and varieties (S4 Table). As the data suggested that variant b3 is private to the Ragdoll population, blood type frequencies for each breed were estimated considering variants b1, b2 and c genotype (available across the entire study sample) for non-Ragdoll breeds and including b3 genotype results for the Ragdoll breed only. Based on genetic blood type determination, blood type B was most common in the following five breeds or breed groups with >15 individuals tested: American Curl (40.4%), British Shorthair breed types (20.3%), Cornish Rex (33%), Devon Rex (30.3%) and Havana Brown (20%). In addition, the breeds in which the rare blood type AB was present with a frequency of >1% were European Shorthair (2.2%), Lykoi (1%), Scottish Fold (3.3%), RagaMuffin (3.4%), Ragdoll (2.2%) and Russian Blue breed types (1.5%). The proportions of type A, type B and type AB blood for each breed or breed group with >15 individuals tested are shown in Table 3 (breed groups listed in S6 Table). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 3. Genetically determined proportions of type A, type B and type AB blood for each breed or breed type with >15 individuals tested. https://doi.org/10.1371/journal.pgen.1009804.t003

Factor XII Deficiency and Pyruvate Kinase Deficiency are widespread blood disorders in the cat population Factor XII Deficiency is a widely distributed heritable trait in the domestic cat population [38]. Factor XII Deficiency is a clinical hemostatic defect that manifests as a prolonged activated partial thromboplastin time (aPTT) which would be observed in a presurgical coagulation assay, but does not require transfusions [39]. Two variants of the F12 gene: c.1321delC and c.1631G>C, have been identified in a colony of inbred cats from the United States and in a litter of cats from Japan, respectively [40,41]. These variants are both considered common in the domestic cat [39]. The most severe aPTT prolongation is observed in cats homozygous for both variants [39]. In accordance with previous observations [39], we noted that the c.1321delC variant always co-segregates with c.1631G>C, while one or two copies of the latter can be inherited in the absence of the c.1321delC variant. The observed variant frequencies for the c.1321delC and c.1631G>C variants were 1.3% and 7% in all cats (Table 1). The frequency of the c.1631G>C variant was based on a subset of 2,186 genotyped cats, as the variant represented a more recent discovery added to the genotyping platform. The presence of the tested F12 variants was found in the following breeds in which cases of clinical Factor XII Deficiency have been documented: Himalayan, Maine Coon, Manx, Munchkin, Oriental Shorthair, Persian, Ragdoll, Siberian, and Siamese. This supports the tested variants’ causal role in Factor XII Deficiency. Moreover, we discovered the presence of the c.1631G>C variant in the Turkish Angora, which is related to the Turkish Van breed, in which clinical cases of Factor XII Deficiency have been documented [39], but no representatives of this breed were in the subset of 2,186 cats screened for the presence of the c.1631G>C variant. Additionally, the tested F12 gene variants were absent in Norwegian Forest Cats, in which the recently identified candidate variant c.1549C>T of F12 gene could potentially be a more common cause of observed cases of Factor XII Deficiency [39]. In all, we identified the two variants of the F12 gene present in 13 more breeds and breed types, and the c.1631G>C variant present alone in 19 additional breeds and breed types. The Tennessee Rex represents a breed with high frequency (>40%) for both variants observed together (Table 2). Finally, we obtained laboratory results from a 10-month-old intact female Maine Coon homozygous for the c.1321delC and c.1631G>C variants of F12 gene. This individual showed a prolonged aPTT of >180 seconds (laboratory reference <13.4 seconds), confirming Factor XII Deficiency in a Maine Coon and further evidence of association between the tested variants and clinical signs. Pyruvate Kinase Deficiency (PK-def) is an inherited anemia characterized by low levels of the pyruvate kinase enzyme. The insufficient presence of the pyruvate kinase enzyme causes red blood cells to break easily, resulting in hemolytic anemia. PK-def shows a marked clinical variability, including variation in the age of disease onset and severity [42,43]. A single nucleotide substitution (c.693+304G>A) in intron 5 of the PKLR gene, with a hypothesized impact on splicing, has been associated with the manifestation of PK-def in the Abyssinian and Somali breeds [43], but has also been previously reported in at least 15 additional cat breeds. Across the entire dataset the PK-def associated variant was found at 3.1% and 2.2% in the Abyssinian and Somali breeds respectively, and was also present in 21 additional breeds (Tables 2 and S4). Several cats from the Bengal, Maine Coon, and Maine Coon Polydactyl breeds were identified genetically at risk with two copies of the PK-def associated variant. Our veterinarians interviewed owners of ten Maine Coons (including two Maine Coon Polydactyls) and three Bengal cats. All the Maine Coons were 3 years old or younger at the time of interview. In three out of ten cases owners reported occurrence of at least one mild potential episode with clinical signs such as lethargy, anorexia, weight loss, and/or jaundice (S7 Table). One male Maine Coon, at the age of 1 year and 5 months, become severely ill with anorexia, lethargy, and significant weight loss (down to approximately 8 lbs from 17 lbs), another male had had episodes of mild lethargy and hyporexia, and the one female cat may have potentially manifested mild symptoms directly after giving birth. The three Bengal cats were between 2-years-9-months-old and 6-years-7-months-old at the time of interview. The owner of one female cat reported that the cat had stopped eating and seemed sick after an attempt to rehome, but fully recovered after returning to the breeder (S7 Table). We also interviewed the owner of a genetically affected Abyssinian cat. The 2-year-5-month-old male cat had four major episodes of illness each time starting with hyporexia and lethargy followed by anorexia, anemia, fever and jaundice, and each time had received significant veterinary care (S7 Table). Compared to the Abyssinian, all owner-reported episodes in the Maine Coon and Maine Coon Polydactyl are milder and should be considered preliminary results as they lack a complete blood count taken during the episode and PK-def diagnoses by a veterinarian.

Autosomal dominant disease-associated variants are observed in additional breeds We screened for four feline disease-associated variants that most closely follow an autosomal dominant mode of inheritance in clinical settings: Polycystic Kidney Disease (PKD) [14], two Hypertrophic Cardiomyopathy (HCM) variants [44,45], and Osteochondrodysplasia and Earfold [46]. Polycystic Kidney Disease (PKD) is a severe autosomal dominant (homozygous lethal) condition in which clusters of cysts present at birth develop in the kidney and other organs, causing chronic kidney disease which can lead to kidney failure [47,48]. PKD is caused by a stop codon in exon 29 of PKD1;c.9882C>A (published as c.10063C>A, coordinates updated to reference genome FelCat9), resulting in a truncated form of the gene, which was discovered in Persian cats with ~40% frequency in the Persian cat population worldwide [16–19]. Genetic testing was introduced into the breeding programs of Persians and some Persian-related cats. Our findings indicate that the overall frequency of the PKD1 variant has decreased notably from what was previously reported in these breeds (Tables 2 and S4). However, the PKD1 variant was identified in the Maine Coon, a breed in which it had not been previously documented in the peer-reviewed literature. Clinical manifestation of PKD in a genetically affected female Maine Coon, diagnosed at the age of 3 months, was confirmed after interviewing the cat’s owner and assessing associated diagnostic documentation including an ultrasound of the kidneys in which numerous, round, well-defined cysts were observed bilaterally throughout the renal cortex and medulla (S7 Table). This finding provides evidence that genetic screening for the PKD1 variant in the Maine Coon is clinically relevant. Hypertrophic cardiomyopathy (HCM) is the most common heart disease in domestic cats. Two independent variants of the MYBPC3 gene c.91G>C p.(A31P) and c.2453C>T p.(R818W) (published as 2060C>T p.(R820W); coordinates updated to reference genome FelCat9) have been associated with HCM in the Maine Coon and Ragdoll breeds, respectively [44,45,49]. In the heterozygous state, the likelihood of developing clinical HCM early in life is very low. However, supporting the autosomal dominant mode of inheritance, regional diastolic and systolic dysfunction has been observed in heterozygous asymptomatic cats [50,51]. In the homozygous state, the development of HCM is highly likely in the Maine Coon with risk increasing with age [52]. Similarly, in the Ragdoll, heterozygous cats have a normal life expectancy, while homozygous cats are likely to have a shortened life span [53]. In the present study, we found the A31P and the R818W variants present in the heterozygous state in additional breeds (Table 2). Our veterinarians interviewed owners of four American Bobtail cats heterozygous for the R818Wvariant. One male had died suddenly at 5 to 6 years of age. Death was preceded by labored breathing over a few days, which the owner suspects was caused by a cardiac condition. However, no echocardiography had been applied to confirm a diagnosis of HCM (S7 Table). All three female American Bobtails at ages 4-years-and-4-months, 6-years-and-5-months and 10-years-and-5-months were still alive. Osteochondrodysplasia and Earfold is a highly penetrant autosomal dominant condition caused by a missense variant (c.1024G>T) in the TRPV4 gene resulting in congenital degenerative osteochondrodysplasia or “Scottish Fold Syndrome”, manifesting as skeletal deformities such as a short, thick, inflexible tail and malformation of the distal fore- and hindlimbs, which can lead to a stilted gait [46]. We observed one copy of the TRPV4 variant in all 85 phenotype-confirmed Scottish Fold cats, and the TRPV4 variant was absent in all 75 Scottish Straight cats tested. We also discovered one copy of the TRPV4 variant in a crossbred cat resulting from the mating of a Scottish Fold and a Highlander. As the ear phenotype of the kitten was curled-back as seen in the Highlander breed, rather than folded forward as seen in the Scottish Fold, observation of the TRPV4 variant was not entirely expected. However, the kitten did present a stiff and inflexible shortened tail, characteristic of TRPV4 variant carriers. It therefore would appear that the yet unknown variant that causes the Highlander ear type masks the Scottish Fold ear phenotype caused by the TRPV4 variant when the two variants are inherited together. In another recent study, a cat registered as an American Curl with curled ears was diagnosed with osteochondrodysplasia and genotypically showed one copy of TRPV4 variant [54]. Thus, we report a second case of Osteochondrodysplasia in which the cat’s ear phenotype belied the presence of the causal variant.

Molecular heterogeneity of feline hereditary retinal dystrophies The disease-associated variants for retinal dystrophies screened in this study include CEP290, KIF3B and AIPL1. The CEP290 variant is associated with late-onset Progressive Retinal Atrophy (PRA; discovered in the Abyssinian) and is present in many pedigreed breeds [55,56]. Here we document a frequency of 1.1% in all cats and have identified the presence of the CEP290 variant in 20 additional breeds. The highest CEP290 variant frequencies were observed in the Peterbald (26.5%) and in one of the additionally identified breeds, the Oriental Longhair (19.6%). The variant of the KIF3B gene, recently associated with an early-onset PRA; discovered in the Bengal) [57], was present in 6.9% of the Bengal breed (resulting in a frequency of 1.1% in all cats) (Tables 1 and 2 and S4). This variant was additionally discovered in the Highlander breed types and the Savannah. The AIPL1 variant associated with PRA (discovered in the Persian) was the rarest variant associated with retinal dystrophies [58]; this variant was screened in a subset of 2,186 samples (including 5 Persian cats) and not observed at all (Tables 1 and S4). To pursue clinical validation of our findings, we recruited a 10-year-3-month-old female Oriental Longhair, homozygous for CEP290. Clinical validation was initiated with an owner interview, in which the owner reported no apparent changes in the cat’s behavior that were suggestive of vision loss (S7 Table). However, during an ophthalmic examination, a marked discoloration of pigmentation of the tapetal fundus with a slight vascular attenuation was noted, confirming the presence of retinal degeneration. This evidence suggests that rdAc-PRA may manifest clinically in the Oriental Longhair, and suggests vision may be retained longer than the previously reported 3–7 years [2], at least for this particular breed example.

Feline MDR1 Medication Sensitivity associated with adverse medication reactions in the Maine Coon Feline MDR1 Medication Sensitivity is a disorder associated with severe adverse reactions after exposure to medications that use the p-glycoprotein drug transporter. This genetic condition is caused by a two base pair deletion within exon 15 of the ABCB1 gene resulting in abnormal p-glycoprotein [59]. While functional p-glycoprotein plays a significant part in the blood-brain barrier that prevents various drugs and chemicals in the bloodstream from entering the brain, a defective p-glycoprotein allows more drugs to cross this barrier, thus increasing the neurological effects of some medications. Severe macrocyclic lactone-induced neurologic toxicosis has previously been reported in cats homozygous for the MDR1 variant receiving either a subcutaneously administered dose of ivermectin or a topically administered eprinomectin-containing antiparasitic product labeled for cats [59,60]. We report the frequency of the ABCB1 variant as 0.6% of all genotyped cats, in addition to the discovery of the variant in the Balinese, Maine Coon, Maine Coon Polydactyl, Ragdoll, Siamese and Turkish Angora breeds (Tables 1 and 2 and S4). Our veterinarians interviewed three owners of cats identified as homozygous for the MDR1 variant to assess their cat’s medical history (S7 Table). The cats consisted of a 1-year-4-month-old intact female Maine Coon, a 2-year-3-month-old intact female Ragdoll, and a 3-year-2-month-old intact male Maine Coon. All three cats had been administered topical flea medications (of varying brands) with no discernable side effects, however, none of the medications applied contained eprinomectin. One of the cats had undergone anesthesia and fully recovered, but reportedly was a bit more lethargic than expected.

Genetic diagnosis plays a crucial role in the diagnosis of uncommon inherited disease We identified a cat genetically affected with Myotonia Congenita, an uncommon recessively inherited disorder manifesting as an inability of the muscles to relax after contraction, which is caused by a variant in the CLCN1 gene [37]. This is a sporadic condition that was discovered in a rescue domestic cat population in Winnipeg, Canada. While the variant was not identified in any pedigreed cats (which make up a large proportion of the study sample), we discovered two copies of the variant in a single non-pedigreed domestic cat in Oregon, United States. In the owner interview, we learned that the genetic diagnosis was crucial in assisting with clinical diagnosis (S7 Table). While this is an incurable condition, having the correct diagnosis helps ensure that the cat is getting appropriate supportive care. The owner confirmed that this cat, initially misdiagnosed with flea bite anemia at the age of 13 weeks, has a disease manifestation that includes fainting spells when startled, prolonged prolapse of the nictitating membrane, hypertrophic musculature, flattened ears, motor dysfunction, shortened gait, and limited range of motion in the jaw. The cat also shows a characteristic “smile” after a yawn or a meow due to delayed relaxation of the muscle in the upper lip as well as commonly has the paws protracted (Fig 1A). However, the owner mentioned that this cat, currently 4 years and 2 months old, is not drooling or showing any dental defects, which differs from the original disease description [37]. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. A) The signature “smile” of a cat with Myotonia Congenita; B) The rare coat color phenotype Amber in a random-bred cat from Finland; C) Polydactyly variant Hw also associated with extra toes in all four feet. Photo credit (from A to C): Kimberly Sullivan, Ari Kankainen, Samantha Bradley. https://doi.org/10.1371/journal.pgen.1009804.g001

Panel screening enables dissociation of variants with clinical disease In this study we determined the c.140C>T p.(S47F) variant of the UROS gene to be present in 0.7% of cats across the entire study sample (Tables 1 and S4). This variant was previously discovered along with c.331G>A p.(G111S) in a cat manifesting Congenital Erythropoietic Porphyria (CEP) [61]. The manifestation of CEP includes distinctively stained brownish-yellow teeth that turn fluorescent pink under UV light. However, we did not observe the c.331G>A variant in any of the tested cats. While the original research revealed no cats carrying solely the c.140C>T variant, functional studies have shown that c.140C>T alone does not significantly alter the protein function [61]. Due to the high frequency of c.140C>T in some cat breeds and because some DNA testing laboratories offer tests for the two variants separately, our veterinarians reached out to the owners of cats homozygous for only the c.140C>T variant resulting in five cases: a 1-year-4-month-old intact female RagaMuffin, a 1-year-5-month-old intact male RagaMuffin, a 1-year-7-month-old intact female Siberian, a 2-year-10-month-old intact female Singapura and a 3-year-7-month-old intact male Toybob. None of these cats had clinical signs suggestive of CEP. Thus, the existing evidence strongly suggests that c.140C>T is a benign variant when it is not inherited along with c.331G>A. Mucopolysaccharidosis Type VI (MPS VI), a lysosomal storage disease caused by a deficiency of N-acetylgalactosamine-4-sulfatase (4S), is another disease in which the roles of two variants, independently inherited in this case, have been under discussion [62]. While the MPS VI variant c.1427T>C p.(L476P) of the ARSB gene is associated with severe disease [63], the c.1558G>A p.(D520N) variant of the ARSB gene is sometimes referred to as the “mild” type [64]. However, it is necessary for the D520N variant to be inherited as a compound heterozygote with one copy of the L476P variant for the disease to manifest in its mild form. We have elected to call the D520N variant MPS VI Modifier. The D520N variant is present in large numbers of cats, with a frequency of 2.5% (Tables 1 and S4). It is also the major allele in the Havana Brown, with an allele frequency of 76.7%. Additionally, we confirm that L476P is a scarce variant, and this study cohort did not identify any cats with MPS VI variant in concordance with previously described observations [62]. The high variant frequency of the benign MPS VI Modifier and very low frequency of the MPS VI variant (absent in many breeds) further confirm that breeding to avoid MPS VI should concentrate solely on managing the L476P variant.

Appearance-associated variants distributed across breeds and breed types In this study sample of mostly pedigreed cats (94.4%), the ancestral form of the appearance-associated (trait) variant was the major allele, except for the ASIP gene in which the derived allele a (non-agouti/solid color) [65] showed a 56.2% variant frequency in all cats (S4 Table; include all within breed variant frequencies). The rarest derived allele observed was the e allele of the MC1R gene associated with the coat color Amber (discovered in the Norwegian Forest Cat) [66], which was observed in the Norwegian Forest Cat and one random-bred cat from Finland. The random-bred cat was homozygous for the derived allele with pictures confirming the expression of the Amber coat color (Fig 1B). Expectedly, a high number of breeds and breed types shared the same trait-associated variants in common (S4 Table). For example, the c.904G>A variant of the TYR gene resulting in Colorpoints (discovered in the Siamese) [67], is also the cause of the Colorpoint phenotype in the Birman, Himalayan (a colorpoint type of Persian cats) Ragdoll, Neva Masquerade (a colorpoint type of Siberian), Ragdoll and Toybob. The derived variant is also present in >30% of the representatives of American Curl, Bengal, British Longhair, Cornish Rex, Donskoy, Highlander, LaPerm, Minuet, Munchkin, Oriental Longhair, Oriental Shorthair, Peterbald, Sphynx and Tonkinese. Genotyping results of trait-associated variants showed overall concordance with observed phenotypes, strongly supporting the variants’ causality. For the two variants White Spotting and Dominant White, which are a result of a full or partial feline endogenous retrovirus (FERV1) insertion into KIT gene [68], we had a genotyping assay that detected if either one of the variants was present requiring combination of genotype with the cat’s phenotype for result interpretation. The White Spotting variant, which we believe to be more ancestral of the two, was much more common in cats than the Dominant White phenotype. In the Turkish Van breed (the name for the breed comes from the “Van pattern” where the colored areas are restricted to the area around the ears with a few additional spots), all the cats had two copies of the White Spotting based on the phenotypic observations. In other breeds with phenotypes, both one and two copies of the White Spotting manifested as a variable coverage of white areas on the cat and based on the cat’s appearance, it was not possible to determine how many copies of the White Spotting was present. However, we observed that in cats with white paws (even when the rest of the legs had color), at least one copy of the White Spotting was present in all breed backgrounds. The only exception were the Birman cats, in which the breed-defining white paws (aka Gloves) are associated with the two adjacent missense changes of the KIT gene [69]. This derived variant had a frequency of 95.6% in the Birman breed and the cats manifested white paws in the absence of White Spotting. Interestingly, we also observed the derived variant as a minor allele in many breeds, in which white paws are not seen. We identified two copies of the Gloves variant present in individuals of the Chartreux, Highlander, Maine Coon, Ragdoll and Siberian breeds, in which they were not associated with Gloving based on photographic evidence and owner reports. Of the phenotyped individuals some white areas of hair were seen on the belly or toes in (3/17) Maine Coons and (1/2) Siberian cats. Our findings are supportive of a potential additional candidate locus that may play a role in the regulation of the Gloves phenotype [70]. Moreover, we observed the Hairlessness variant (discovered in the Sphynx) [71], the Shortened Kinked Tail variant (discovered in the Japanese Bobtail) [72] and the three tested Short tail variants (discovered in the Manx cat) [73] result in their expected phenotypes, but also not explain all observed bald and short tail phenotypes. The c.816+1G>A variant in the KRT71 gene which is known to result in the hairless phenotype of the Sphynx cat had a variant frequency of 74.7% in the Sphynx breed [71] (S4 Table). Compound heterozygotes for the Sphynx c.816+1G>A variant with the Devon Rex associated curly coat variant are also hairless [71]. We also identified 22 hairless Sphynx cats without any copies of the Hairlessness variant, suggesting that an additional unknown variant in the same or another gene entirely causes the hairless phenotype in some cats of this breed. We also confirmed that the Hairlessness variant was not present in the Donskoy and Peterbald breeds, which also represent hairless phenotypes. It was recently suggested that a novel 4 base pair variant of the LPAR6 gene identified in a Peterbald cat potentially results in a hairless coat phenotype as a homozygote or as a compound heterozygote with the c.250_253_delTTTG variant in the LPAR6 gene which causes the curly coat of the Cornish Rex [74]. We show a variant frequency of 25% for the Cornish Rex coat variant in the Donskoy, but did not identify any Cornish Rex coat variant carriers among the 17 tested Peterbald cats of this study. The c.5T>C variant in the gene HES7 associated with a shortened and kinked tail in the Japanese Bobtail [72] was observed as homozygous in all Japanese Bobtail cats of this study, with the derived variant also prevalent in the Kurilian Bobtail and Mekong Bobtail. In additional bobtail breeds, Cymric and Toybob, a few individuals were also observed with the derived variant (S4 Table). All studied cats with one or more copies of HES7 variant and available phenotypic information presented with a shortened tail. We genotyped three out of four known variants (c.995delT, c.1166delC, c.1196delC) (published as 998delT, c.1169delC and c.1199delC; coordinates updated to reference genome FelCat9) of the T gene (discovered in the Manx) causing shortened tail [73]; the fourth T variant (c.995_1011dup;1011_1014del on FelCat9 reference genome, published as 998_1014dup17delGCC), which has only been previously observed in one cat, could not be genotyped. We found one of the three tested variants in all the Pixiebobs (including longhair variant), 9/30 (30.0%) Manx and 7/14 (40.0%) Cymric (a Manx type with longhair). The derived variants of the T-box were always observed in the heterozygous state, further suggesting that in the homozygous state these variants are lethal in utero [73], and therefore in the mating of two short-tailed cats 25% of cats conceived are born with longtail. Long-tailed Manx and Cymric cats are part of the breeding programs and because there was no phenotype information available for all tested Manx and Cymric cats, it could not be estimated if the tested three variants explained all short tail phenotypes in these two breeds. In addition, in line with previous study [73], the variant c.995delT of the T gene was also observed in American Bobtails (combined group of both Longhair and Shorthair variants). The tested Manx shorttail variants were also remarkably common in the Highlander breed, in which there is believed to be a different cause for the short tail phenotype. In this study cohort, 40.6% (94/231) of the Highlanders had one of the tested short tail variants (discovered in the Manx) (S4 Table). The photographic evidence in the American Bobtail and Highlander breeds confirm several short-tailed cats that were not carrying any known bobtail variant (S4 Table). Lastly, we found that of the three tested polydactyly (extra toes) associated variants, the Hw variant of the LMBR1 gene was predominantly observed in polydactylous cats, including the Maine Coon, Pixiebob and Highlander breed types, and some non-pedigreed cats from North America. Both heterozygous and homozygous cats observed, and based on the photographic evidence presenting with four (normal number of digits) to seven toes per paw. Higher variant penetrance was seen in homozygous cats, which is in line with previous observations [75]. However, extra toes did not manifest solely in the front feet as previously reported; photographic evidence and owner-provided details revealed the presence of extra toes on all four feet (Fig 1C), which was formerly considered to be characteristic of the UK1 and UK2 variants only [75]. We also confirmed that a non-pedigreed cat with two copies of the UK1 variant had two extra digits on each paw, per the owner’s description. Various owner-reported cases of polydactyl cats testing negative for the screened variants were also noted. Such cats are likely to be carrying additional variants of the LMBR1 gene or other not yet identified locus.

[END]
---
[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1009804

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/