inherited neurologic disorders in the dog
TRANSCRIPT
Inherited NeurologicDisorders in the Dog
The Science Behind the SolutionsCathryn Mellersh, BSc, PhD
KEYWORDS
� Inherited disorder � DNA testing � Genetic mutation
KEY POINTS
� Inherited neurologic diseases are varied and can be congenital, neonatal, or late onset aswell as progressive or stationary.
� Modern genetic technologies are revolutionizing the speed and efficiency with whichmutations responsible for inherited neurologic disease are being identified.
� Clinically similar disorders can be caused by different mutations, even within a singlebreed, and are thus genetically distinct.
� DNA tests can be used by dog breeders to reduce the prevalence of inherited neurologicdisorders in specific breeds and help the veterinarian diagnose disease.
BROAD CHARACTERISTICS OF INHERITED DISEASE—WHEN SHOULD A VETERINARIANSUSPECT A DISEASE IS INHERITED?
There is no definitive or trademark characteristic of an inherited disorder, and theveterinarian should always be open minded about the possibility that a patient maybe suffering from an inherited condition. A variety of inherited neurologic diseaseshas been described in the dog, including examples that are congenital, neonatal,and late onset as well as those that are progressive and stationary, so potentiallyany neurologic patient could be suffering from an inherited disorder.It is not possible to tell whether a disease is inherited from a single case; the only
indication a disease might be inherited is whether other dogs of the same breed orfrom the same extended pedigree have also been reported with the same or similarclinical presentation. Mutations that cause inherited disorders arise at random infounder animals and are passed to offspring and subsequent generations if thefounder reproduces. If the mutation is recessive, clinically affected animals areonly produced when inbreeding has occurred and a dog inherits an identical
Centre for Preventive Medicine, Animal Health Trust, Lanwades Park, Kentford, Newmarket,Suffolk CB8 7UU, UKE-mail address: [email protected]
Vet Clin Small Anim 44 (2014) 1223–1234http://dx.doi.org/10.1016/j.cvsm.2014.07.011 vetsmall.theclinics.com0195-5616/14/$ – see front matter � 2014 Elsevier Inc. All rights reserved.
Mellersh1224
copy of the mutation from both parents. For this reason, specific inherited diseasesare nearly always associated with particular breeds or several closely relatedbreeds.
Sources of Data Regarding Inherited Diseases in Domestic Animals
A literature review might reveal whether a specific disorder has been reported previ-ously. PubMed1 is a free search engine accessing primarily the MEDLINE databaseof references and abstracts on life sciences and biomedical topics and is a sensibleplace to initiate a search for evidence a disease might be inherited.Another source of evidence that a disorder may be inherited in a domestic spe-
cies is Online Mendelian Inheritance in Animals (OMIA), which is a catalog/compen-dium of inherited disorders, other (single-locus) traits, and genes in 214 animalspecies (other than human and mouse and rats, which have their own resources)authored by Professor Frank Nicholas of the University of Sydney, Australia.2
OMIA information is stored in a database that contains textual information and ref-erences as well as links to relevant PubMed (described above) records at the Na-tional Center for Biotechnology Information and to the equivalent database forhuman inherited disorders, Online Inheritance in Man,3 and to Ensembl,4 a softwaresystem that produces and maintains automatic annotation on selected eukaryoticgenomes.A database that compiles information specifically about diseases/conditions of
pure-bred dogs that are likely to have a genetic component is the Inherited Diseasesin Dogs database compiled by Dr David Sargan at Cambridge University.5
In addition to online sources of information, an excellent text describing canine andfeline disorders that are potentially breed associated is Breed Predispositions to Dis-ease in Dogs and Cats, 2nd Edition by Alex Gough and Alison Thomas,6 which in-cludes reference to a peer-reviewed publication for each disorder described.However, not all inherited diseases will have been described in the scientific litera-
ture, although considerable anecdotal evidence might still exist to suggest a diseasehas an inherited component. Breed clubs and breed societies as well as individualbreeders are frequently knowledgeable regarding conditions that segregate in theirbreed, often well before they have been formally recognized by the veterinary profes-sion, so making contact with a breed health coordinator or the equivalent might alsoyield useful information.
IDENTIFYING THE UNDERLYING CAUSE OF AN INHERITED DISORDER
Once it has been established that a specific disorder is likely to be inherited, by virtueof the fact that it is more prevalent in certain breeds than others, it becomes desirableto identify the mutation(s) or genetic variant(s). Once the causal mutation is known, aDNA test can be developed that breeders can use to guide their breeding decisionsand reduce the prevalence of the condition in their breed. The opportunity to minimizethe risk of producing clinically affected puppies is particularly desirable when the dis-order is challenging to treat effectively or is particularly debilitating—as is frequentlythe case for diseases of the nervous system.Over the last decade, the tools available to dissect the genetic basis of canine
inherited traits have become increasingly sophisticated since the canine genomewas sequenced in its entirety in 2004. The current rapid rate at which disease muta-tions are identified can be expected to increase further in coming years as next-generation sequencing techniques become increasingly cost effective and thereforewithin the reach of even the most modestly sized research groups.
Inherited Neurologic Disorders in the Dog 1225
Methodology of Mutation Identification
There are 2 main routes to mutation identification; the genomewide investigationapproach and the candidate gene approach. During a genomewide investigation,the entire genome is investigated for region(s) likely to harbor the disease-associated mutation(s), using pedigree-based linkage analysis, a genomewide asso-ciation study (GWAS), or whole genome sequencing. In contrast, during a candidategene study, only specific genes are investigated. The successful outcome of agenomewide linkage or association study is the identification of a genomic regionthat is likely to contain the causal mutation being sought. In-depth investigations ofthe associated region follow, which might involve analysis of candidate genes withinthe region or a more holistic sequencing approach (known as targeted resequencing,see later discussion). In whole genome sequencing, the entire genome of an affectedanimal is sequenced and investigated for pathogenic mutations. The different routesto mutation identification are summarized in Fig. 1 and described later.
Genetic Markers
Both linkage and genomewide association analyses use genetic markers to identifygenomic regions associated with disease. Genetic markers are variable or polymor-phic regions of DNA that help geneticists navigate their way around the genomeand identify regions of DNA that are associated with traits of interest. For geneticmarkers to be useful, their positions relative to the genome and their positions relativeto one another must be known. Microsatellites are one commonly used form of geneticmarker that comprise a simple DNA motif repeated in tandem a variable number oftimes. Often, these repeats consist of the nucleotides, or bases, cytosine and adeno-sine—so-called CA repeats. The number of repeats present within a given microsat-ellite may differ between individuals, hence, the term polymorphism—the existenceof different forms within a population. Single nucleotide polymorphisms (SNPs) are
Fig. 1. Schematic illustration of different potential routes to mutation identification.
Mellersh1226
individual point mutations, or substitutions of a single nucleotide, that do not changethe overall length of the DNA sequence in that region. Both SNPs and microsatellitesoccur frequently and regularly throughout an individual’s genome, making them idealtools for genetic mapping. The alternative forms of a gene or genetic marker thatoccupy the same locus on a chromosome are known as alleles.
Genetic Linkage Analysis
Markers, genes, or mutations that are located on the same chromosome as oneanother tend to be inherited together, whereas markers on different chromosomesare inherited independently. The closer 2 markers are to each other the less likelythey are to be separated by genetic recombination, the process that occurs duringmeiosis when chromosome pairs align and physical breakage and exchange of ge-netic materials between the homologous chromosomes can occur. Recombination re-sults in a combination of genes/markers different from that of either parent. Byanalyzing large numbers of genetic markers from all over the genome (a genomewidescan) in dogs from pedigrees segregating an inherited condition, researchers canidentify markers that are being co-inherited with the disease. Because the locationof genetic markers within the genome is known, the position of the disease-causingmutation can also be inferred. The above process is known as pedigree linkage anal-ysis and requires the analysis of DNA from both affected and unaffected related mem-bers of an extended family. The precision with which a mutation can be locateddepends on several factors, including the numbers of animals in the study, thenumbers of genetic markers analyzed, and the fortuitous occurrence of recombinationin critical dogs within the pedigree.
ExampleIn practice, it is often difficult to collect DNA samples from sufficient family membersfor a successful linkage analysis unless a research colony is available. However, thetechnique was used very successfully in 1999 by Lin and colleagues7 to identify a mu-tation in the hypocretin (orexin) receptor 2 gene (Hcrtr2) that was responsible for awell-established canine model of narcolepsy.
Genomewide Association Studies
An alternative approach to linkage analysis is to use a GWAS, also known as associ-ation mapping, whereby unrelated affected and unaffected individuals (cases andcontrols) are drawn from the population, and the frequency with which certain allelesare present in each of these groups is tested for association with a disease. Similar tolinkage analysis experiments, the successful outcome of an association study is theidentification of a region of the genome containing a mutation that is responsible forthe disease under investigation. Because modern breeds of dog have been developedrelatively recently, often from a small number of founders, individuals of the samebreed exhibit low levels of intrabreed variation and usually share long regions of chro-mosomes (haplotype blocks) that are identical by descent. This extensive linkagedisequilibrium is in contrast to that in humans and means that far fewer, less denselyspaced markers are sufficient to map genomic regions associated with traits in thedog. The disadvantage of long linkage disequilibrium is that although it is relativelyeasy to map disease regions, the regions are long, relative to corresponding regionsthat are mapped in more genetically diverse species. But once a broad region of inter-est has been identified, the geneticist can take advantage of the high levels of inter-breed variation. Different breeds are genetically isolated from one another, meaningthe genetic profiles of different breeds are distinctive. Traits shared by different breeds
Inherited Neurologic Disorders in the Dog 1227
can be linked to large genomic regions using a single breed, and then the associatedregion can be refined and reduced by identifying theminimum haplotype that is sharedacross breeds segregating the same trait. Although each breed studied will carry thegene of interest on a large haplotype block, each breed’s haplotype will be different,having arisen from an independent set of recombination events; the overlapping re-gion that is common to all breeds contains the gene being sought.
ExampleAn example of the successful use of a GWAS is that of Forman and colleagues8 whoused this method to successfully identify the mutation for spinocerebellar ataxia (SCA)in the Parson Russell terrier (PRT). SCA in the PRT is a disease of progressive incoor-dination of gait, and loss of balance and pedigree analysis indicated an autosomalrecessive mode of inheritance. Clinical signs usually become apparent between 6and 12 months of age with affected dogs presenting with symmetric SCA particularlyin the pelvic limbs. The degree of truncal ataxia, pelvic limb hypermetria, and impairedbalance is progressive, particularly during the initial months of disease. A certain de-gree of stabilization and intermittent worsening may occur. At the later stages of thedisease, ambulation often becomes difficult, with owners usually electing to euthana-tize affected dogs on welfare grounds. Using a GWAS approach followed by targetedresequencing (see later discussion), a SNP in the CAPN1 gene, encoding the calcium-dependent cysteine protease calpain1 (mu-calpain), was identified for which allaffected dogs were homozygous. The SNP is a missense mutation causing a cysteineto tyrosine substitution at residue 115 of the CAPN1 protein. Cysteine 115 forms a keypart of a catalytic triad of amino acids that are crucial to the enzymatic activity ofcysteine proteases. The CAPN1 gene shows high levels of expression in the brainand nervous system, and roles for the protein in both neuronal necrosis and mainte-nance have been suggested. Given the association with SCA in the PRT, the functionalimplications and high level of conservation observed across species, CAPN1 repre-sents a novel potential cause of ataxia in humans.8
Targeted Resequencing
Targeted resequencing is the term used to describe the process of sequencing a smallsubset of the genome, such as a particular chromosome or a chromosomal region ofinterest. The technique is commonly used once a GWAS has implicated a particulargenomic region to be associated with a disease under investigation (see previous dis-cussion). The associated region is typically resequenced in some cases and controlsand their sequences compared to identify any variants that are predicted to be path-ogenic. The term resequencing is used because the method is used to identifygenomic variations of a DNA sample, or small cohort of samples, in relation to a com-mon reference sequence. A detailed description of targeted resequencing methodol-ogy is outside the scope of this review, but briefly, after fragmenting the genome,fragments from the desired region are captured by hybridizing the sample to comple-mentary biotinylated probes, which can then be separated from the rest of the genomeusing streptavidin-labeled magnetic beads, followed by a wash step to remove un-bound, nontargeted fragments. The resulting DNA is then used to prepare a standardlibrary for sequencing.
ExampleThe method was used to successfully identify the mutation responsible for episodicfalling (EF) in the Cavalier King Charles spaniel.9 In this study, a GWAS and targetedresequencing of DNA from just 5 dogs were used to simultaneously map and identify
Mellersh1228
mutations for 2 distinct inherited disorders that both affect the Cavalier King CharlesSpaniel. The authors investigated EF, a paroxysmal exertion-induced dyskinesia,alongside the phenotypically distinct condition, congenital keratoconjunctivitis siccaand ichthyosiform dermatosis, commonly known as dry eye curly coat syndrome.EF is characterised by episodes of exercise-induced muscular hypertonicity andabnormal posturing, usually occurring after exercise or periods of excitement. Thecausal mutation for EF was identified as an approximately 16-kilobase (kb) deletionencompassing the first 3 exons of the brevican gene (BCAN). Brevican is one of thecentral nervous system–specific members of the hyaluronan-binding chondroitin sul-fate proteoglycan family. Brevican is important in the organization of the nodes ofRanvier in myelinated large-diameter axons, and disruption of this region results in adelay in axonal conduction.9
Candidate Gene Analysis
Candidate gene analysis is the term used when a gene or subset of genes, as opposedto the entire genome, are investigated for their potential association with a disease un-der investigation. Candidate genes can be identified at the start of a study based on apriori knowledge of the gene’s biological function, an association with a similar dis-ease in a different species, or at completion of a successful GWAS when a genemay be identified as a positional as well as a functional candidate.
ExampleA good example of a study in which a disease-associated mutation was identified in agene that was both a functional and positional candidate was that of Zeng and col-leagues10 who used this approach to identify the mutation for Bandera’s NeonatalAtaxia (BNAt) in the Coton de Tulear breed of dog. Dogs affected with this condition,which is named after one of the first puppies to be clinically evaluated, are usuallyidentified as soon as their littermates develop coordinated movements. They exhibittitubation of the head and intention tremors. Most are unable to walk but can scootin sternal recumbency as a means of purposeful locomotion. Spinal reflexes remainintact, but righting reflexes are delayed, and proprioceptive positioning is severelydecreased or absent. Affected puppies are visual but lack a menace response andexhibit fine vertical ocular tremors at rest and saccadic dysmetria together with an up-beat nystagmus during dorsal positioning.11 Zeng and coworkers10 undertook aGWAS with 12 cases and 12 controls that identified a 713-kb region on canine chro-mosome 1 that was significantly associated with the disease. The region contained 4genes, one of which was the metabotropic glutamate receptor 1 gene (GRM1) thatwas considered to be the most likely candidate to contain the BNAt-causing mutationbecause naturally occurring and experimentally induced Grm1 deficiencies produceddisease phenotypes in mice that resembled that of the BNAt-affected canine puppies.Resequencing GRM1 from affected dogs identified a 62-base pair (bp) insertion inexon 8 that was concluded to be the cause of BNAt.10
Whole Genome Sequencing
Another route to mutation identification is to compare the sequence of an affectedindividual’s entire genome with those of unaffected individuals. This is known aswhole genome sequencing and is undertaken in the absence of prior associationmapping that can narrow the hunt for the mutation to a small fraction of the genomeand a subset of genes. Whole genome sequencing poses a considerable challenge,from a computational and bioinformatics perspective, because of the vast numberof variants that are typically detected between the genomes of different individuals,
Inherited Neurologic Disorders in the Dog 1229
even when those individuals are members of the same breed of dog. A refinementof whole genome sequencing is to sequence just the coding exons of all genes, atechnique known as whole exome sequencing. This method is less computationallychallenging than whole genome sequencing, because the exome is typically lessthan 5% the size of the whole genome, but relies on the hypothesis that mostmutations associated with disease lie in the coding regions of genes. The methodwill not lead to the successful identification of causal mutations that lie outsidecoding exons.
ExampleWhole genome sequencing has recently been used successfully for the first time in thedog to identify the mutation responsible for a form of inherited SCA associated withmyokymia, seizures, or both known to affect Jack Russell terriers, Parson Russell ter-riers, and Russell terriers (collectively referred to as Russell group terriers).12 The au-thors of this study sequenced the entire genome of a single Russell group terrier withSCA and myokymia and compared it with the whole genome sequences from 81 othercanids that were normal or had other diseases. A mutation in the gene coding for theinwardly rectifying potassium channel Kir4.1 (KCNJ10) was identified that changed anamino acid and was predicted to be pathogenic. All dogs that were homozygous forthis mutation had SCA with varying combinations of myokymia and seizures.12 Thisform of ataxia is different and genetically distinct from the form described abovethat is caused by the CPN1 mutation,8 although both forms are characterized bysimilar and overlapping clinical signs and affect the same breeds.
Whole Transcriptome Sequencing
All of the methods described above use genomic DNA as the substrate for mutationidentification.Analternative to sequencinggenomicDNA is so-calledmRNAsequencingfroma tissue that is central to the trait. BecausematuremRNA is effectively anRNAcopyof the exome, mRNA sequencing can be considered a form of targeted resequencing,allowing nature to do the target capture rather than the in vitromethods that are requiredfor exome sequencing.
ExampleSequencing ofmRNAhas, to date, only been used once to identify a disease-associatedmutation in the dog. Forman and colleagues13 used the method to identify the mutationresponsible for neonatal cerebellar cortical degeneration (NCCD) in the beagle. NCCD-affected beagles are unable to ambulate normally from the onset of walking, and themain pathologic findings include Purkinje cell loss with swollen dendritic processes. Inthis landmark investigation, mRNA sequence from the cerebellum of a single NCCDcase was generated, and the sequence of 27 genes known to cause ataxia in humanswas analyzed. The causal mutation, located in the b-III spectrin gene (SPTB2), was an8-bp coding deletion that is predicted to cause both an aberrant run of 27 extra aminoacids and premature termination of mRNA (Fig. 2). The mutation segregated perfectlyas an autosomal recessive in the small family tested, was found in the heterozygousstate in other unaffected but at-risk dogs and was absent in 37 other breeds. As ex-pected, cerebellar tissue from the affected dog showed a near total loss of b-III spectrinmRNA and protein when compared with a control dog. Spectrins are a family of cyto-skeletal proteins, with tetrameric structures comprising 2 a and 2 b subunits, with diver-sity and specialization of function. b-III spectrin is primarily expressed in the nervoussystem and the highest levels of expression are found in Purkinje cell soma anddendrites.14
Fig. 2. Sequencing data show mutation responsible for NCCD in the beagle. (A) RNAsequencing reads from the mRNA-seq experiment aligned across the deletion and visualizedin the software Integrative Genomics Viewer (IGV).24 Reads are represented by gray bars,with the deletion indicated with a black horizontal line in reads. A single, benign nucleotidepolymorphism (c.5580 T > C) is also located 18 bp downstream of the deleted sequence inthe NCCD case and is highlighted in blue. (B) Sanger sequencing to confirm the 8-bp dele-tion in the case, the sire of the case (obligate heterozygote), and a wild-type individual (sib-ling). The 8-bp sequence upstream of the deletion is identical to the deleted sequence.
Mellersh1230
Inherited Neurologic Disorders in the Dog 1231
DNA TESTING
Once a mutation has been identified, using one or a combination of the methods out-lined previously, a DNA test can be developed and offered to the public. Worldwidethere are now many facilities offering canine DNA tests.15 The process of DNA testingis simple and involves the submission of a sample of a dog’s DNA to an appropriatetesting laboratory. The DNA can often be submitted as a cheek swab that an ownercan take themselves, although some tests or laboratories may require a blood sample.The testing laboratory analyses the DNA for the presence or absence of the relevantmutation and will report back, usually within a few weeks, with the result (the dog’sgenotype). The results will inform the veterinarian or owner whether the dog beingtested has zero, 1, or 2 copies of the mutation for which the dog is being tested.
DNA Tests for Disease-Associated Mutations—What Do the Test Results Mean?
It is worth considering what the results of a DNA test mean to the owner, the veteri-narian, and the dog. The primary consumer of DNA tests is the dog breeder, andthe benefits of DNA tests to breeders wishing to reduce the prevalence of diseasehave been widely discussed.16–19 As increasing pressure is put on breeders toimprove the genetic health of the dogs they produce, the judicious use of DNA testsin this endeavor has never been more topical. It is important for both the breederand the veterinarian to remember that clinically similar conditions can be caused bydifferent mutations, and although clinically affected dogs of the same breed will usu-ally share the same causal mutation, it is possible for genetically distinct forms of thesame disease to segregate within the same breed. A good example of this is illustratedby the 2 clinically similar but genetically distinct forms of ataxia that affect Jack andParson Russell terriers that were described previously. It is important for ownersand veterinarians to appreciate that most DNA tests only assay for a single, specificmutation and not for any other mutations that cause clinically similar conditions. Aclear DNA test result is not, therefore, an absolute guarantee that a dog will neverhave a clinically similar disease to that being for which it is being tested, althoughdogs that are clear of specific mutations can be considered at low risk of diseasedevelopment.
Basic Genetics
Most of the DNA tests currently available are for mutations responsible for simple orsingle gene diseases and include all of the disorders described previously. This meansthat the disease is a result of a single mutation; no other genes or environmental fac-tors are involved. For these diseases, the results of DNA tests are easy to interpret,and an individual dog’s risk of developing the condition can be estimated with ahigh level of certainty from the DNA test results. Many simple inherited conditionshave a recessive mode of inheritance. Recessive diseases are the result of mutationsthat cause the loss of function of a biologically important gene as opposed to domi-nant conditions, which usually result from mutations that cause an inappropriategain of function of a gene. Every dog has 2 copies of each gene, one inherited fromthe dam and one from the sire, and carriers that have inherited a single copy of thenormal gene from one parent and a single copy of a mutant gene from the other parentusually have sufficient functional protein to remain clinically healthy. It is only when adog inherits a faulty gene from both parents that it becomes clinically affected. Conse-quently, if a mutation is recessive, then dogs with zero or 1 copy of the mutation willremain clinically free of the disease, although heterozygous carriers will pass themutation onto around half of their offspring. Dogs with 2 copies of the mutation
Mellersh1232
(homozygotes) will almost certainly have the disease during their lifetime, althoughthey might be clinically clear at the time of testing. If a mutation is dominant, dogswith 1 or 2 copies of the mutation will get the condition (unless there is evidence ofincomplete penetrance), whereas dogs that are clear of the mutation will remainhealthy.Some diseases are more complex, and result from mutations in multiple genes or
the interaction between genes and the environment. Individual mutations might in-crease a dog’s risk of getting the associated condition but cannot predict with cer-tainty whether a dog will become clinically affected.
Using DNA Test Results to Reduce the Prevalence of Disease
DNA tests can play a critically important role in the control and eventual elimination ofinherited diseases. Recessive diseases are notoriously difficult for the dog breeder toeliminate because of the existence of clinically healthy carriers within the populationthat can only be detected retrospectively, once they have produced affected offspringor one of their parents has been diagnosed as affected. The problem is confounded forlate-onset conditions in which affected animals may be innocently bred before theyare diagnosed, and this problem is applicable to dominant and recessive diseases.The availability of a DNA test is often the only way in which a recessive condition or a
late-onset dominant condition can be reliably eliminated from a breed. Breedersshould have their breeding stock tested before mating and make sensible breedingchoices, based on the genotype of their dog, that minimize the risk of producingaffected offspring. Disease mutations can be common within specific breeds, andonce a DNA test becomes available, the instinct of many breeders is to only breedfrom clear dogs. This practice will obviously eliminate the disease mutation from thebreed rapidly but may do so at the expense of genetic diversity if large numbers ofdogs are instantly removed from the gene pool. High levels of inbreeding and lossof genetic variation are well documented to have detrimental effects on the healthand fertility of animals. For common recessive mutations, it is therefore advisablefor breeders to continue breeding with carriers, at least for the first-generation afterDNA test development. Provided all carriers are paired with DNA-tested, clear mates,only clear and carrier puppies will be born; no clinically affected dogs will be produced,and breeders can select a clear dog to breed on from the resulting litters. Table 1 de-tails the outcomes of mating dogs with different genotypes (with respect to a recessivemutation) and whether they can result in clinically affected offspring.
Table 1Outcomes of mating dogs with different autosomal recessive genotypes
Combination of Dogs OutcomePossibility of Clinically AffectedOffspring?
Clear � Clear All puppies will be clear No
Clear � Carrier 50% of puppies will be clear50% of puppies will be carriers
No
Clear � Affected All puppies will be carriers No
Carrier � Carrier 25% of puppies will be clear25% of puppies will be affected50% of puppies will be carriers
Yes
Carrier � Affected 50% of puppies will be affected50% of puppies will be carriers
Yes
Affected � Affected All puppies will be affected Yes
Inherited Neurologic Disorders in the Dog 1233
For dominant mutations, the situation is different. All offspring that inherit a disease-associated dominant mutation will have clinical signs at some stage during their lives,so breeding with animals that carry such mutations is harder to justify.
Using DNA Test Results to Aid Differential Diagnosis
Another, often overlooked, role of the DNA test is to help the veterinarian diagnose dis-ease. There can be considerable overlap between the clinical signs of geneticallydifferent disorders, and DNA tests, where they exist, can come to the rescue. Forexample, several neurologic syndromes have been described in Cavalier King Charlesspaniels, including occipital hypoplasia/syringomyelia, episodic collapse, epilepsy,and vestibular disorders, and clinical signs of these disorders can overlap.20 In 2012,the mutation responsible for one of these conditions, EF, was identified (describedpreviously), and now a DNA test is available for veterinarians to use to assist their inves-tigations for those patients whose clinical signs are consistent with EF.9 Similarly, L-2-hydroxyglutaric aciduria is a neurometabolic disorder that produces a variety of clinicalneurologic deficits, including psychomotor retardation, seizures, and ataxia, and canthus be misdiagnosed as epilepsy. However, since 2007, when the molecular defectresponsible for this condition in Staffordshire bull terriers was characterized, veterinar-ians can determine whether a dog is carrying 2 copies of the causal mutation and there-forewhether it is in fact affectedwith L-2-hydroxyglutaric aciduria or another condition.21
Although a useful aid to differential diagnosis, the results of DNA testing shouldrarely be used in isolation by the veterinarian. One additional factor is the typicalage of onset of specific diseases; inherited disorders typically have a characteristicage of onset, and this should always be considered alongside the DNA test results.One disease in which this is particularly pertinent is degenerative myelopathy (DM),for which a DNA test based on amutation in a gene called SOD1 is available to multiplebreeds.22 The age of onset of DM is variable, ranging from 6 to 15 years of age or evenolder, and some dogs that are homozygous for the DM mutation may die before anyDM signs develop. Therefore, a young dog that is homozygous for the DM mutationand showing clinical signs consistent with DM, may be suffering from an entirelydifferent condition.Some inherited disorders require an environmental trigger or exposure in addition to
carrying a risk mutation. An example is the condition known as exercise-inducedcollapse (EIC) in Labrador retrievers. EIC is an autosomal recessive syndrome causedby a mutation in the DNM1 gene, which causes a defect in nerve communication duringintense exercise.23 Dogs that are homozygous for themutation will collapse, but only af-ter intense exercise, so alternative diagnoses should be considered for dogs that havenot exercisedbefore collapse, regardless of their DNA test result for theDNM1mutation.DNA tests represent a valuable tool with which to reach a differential diagnosis but,
as with all test results, should be considered as part of the complete clinical history fora patient and not in isolation.
Resources listing currently available DNA tests:http://www.akcchf.org/canine-health/health-testing/http://research.vet.upenn.edu/DNAGeneticsTestingLaboratorySearch/tabid/7620/Default.aspx
http://www.thekennelclub.org.uk/media/14688/dnatestsworldwide.pdf
REFERENCES
1. Available at: http://www.ncbi.nlm.nih.gov/pubmed/. Accessed July 2014.2. Available at: http://omia.angis.org.au/home/. Accessed July 2014.
Mellersh1234
3. Available at: http://www.omim.org/. Accessed July 2014.4. Available at: http://www.ensembl.org/index.html. Accessed July 2014.5. Available at: http://www.vet.cam.ac.uk/idid/. Accessed July 2014.6. Gough A, Thomas A. Breed predispositions to disease in dogs and cats. 2nd edi-
tion. Blackwell Publishing; 2010. p. 330.7. Lin L, Faraco J, Li R, et al. The sleep disorder canine narcolepsy is caused by a
mutation in the hypocretin (orexin) receptor 2 gene. Cell 1999;98(3):365–76.8. Forman OP, De Risio L, Mellersh CS. Missense mutation in CAPN1 is associated
with spinocerebellar ataxia in the parson russell terrier dog breed. PLoS One2013;8(5):e64627.
9. Forman OP, Penderis J, Hartley C, et al. Parallel mapping and simultaneoussequencing reveals deletions in BCAN and FAM83H associated with discrete in-herited disorders in a domestic dog breed. PLoS Genet 2012;8(1):e1002462.
10. Zeng R, Farias FH, Johnson GS, et al. A truncated retrotransposon disrupts theGRM1 coding sequence in Coton de Tulear dogs with Bandera’s neonatal ataxia.J Vet Intern Med 2011;25(2):267–72.
11. Coates JR, O’Brien DP, Kline KL, et al. Neonatal cerebellar ataxia in Coton de Tu-lear dogs. J Vet Intern Med 2002;16(6):680–9.
12. Gilliam D, O’Brien DP, Coates JR, et al. A homozygous KCNJ10 mutation in JackRussell Terriers and related breeds with spinocerebellar ataxia with myokymia,seizures or both. J Vet Intern Med 2014;28(3):871–7.
13. Forman OP, De Risio L, Stewart J, et al. Genome-wide mRNA sequencing of a sin-gle canine cerebellar cortical degeneration case leads to the identification of adisease associated SPTBN2 mutation. BMC Genet 2012;13(1):55.
14. Sakaguchi G, Orita S, Naito A, et al. A novel brain-specific isoform of beta spec-trin: isolation and its interaction with Munc13. Biochem Biophys Res Commun1998;248(3):846–51.
15. Available at: http://www.offa.org/dna_alltest.html. Accessed July 2014.16. Mellersh C. DNA testing man’s best friend. Vet Rec 2011;168(1):10–2.17. Mellersh C, Sargan D. DNA testing in companion animals - what is it and why do
it? Practice 2011;33(9):442–53.18. Mellersh C. DNA testing and domestic dogs. Mamm Genome 2012;23(1–2):
109–23.19. Mellersh C. DNA testing: diagnosing and preventing inherited disorders in dogs.
Vet Rec 2013;172(10):264–5.20. Rusbridge C. Neurological diseases of the Cavalier King Charles spaniel. J Small
Anim Pract 2005;46(6):265–72.21. Penderis J, Calvin J, Abramson C, et al. L-2-hydroxyglutaric aciduria: character-
isation of the molecular defect in a spontaneous canine model. J Med Genet2007;44(5):334–40.
22. Awano T, Johnson GS, Wade CM, et al. Genome-wide association analysis re-veals a SOD1 mutation in canine degenerative myelopathy that resembles amyo-trophic lateral sclerosis. Proc Natl Acad Sci U S A 2009;106(8):2794–9.
23. Patterson EE, Minor KM, Tchernatynskaia AV, et al. A canine DNM1 mutation ishighly associated with the syndrome of exercise-induced collapse. Nat Genet2008;40(10):1235–9.
24. Robinson JT, Thorvaldsdottir H, Winckler W, et al. Integrative genomics viewer.Nat Biotechnol 2009;29(1):24–6.