The authors have declared that no competing interests exist.
Conceived and designed the experiments: KV JR DB PD HH. Performed the experiments: KV DB EB HH RJH SD CA JC CG BS AB. Analyzed the data: KV DB EB RJH SD CA JC CG BS AB BR BP GDS. Contributed reagents/materials/analysis tools: DB GDS BR CA JC HH PD RH KV JR. Wrote the paper: DB GDS BR CA JC PD RH CG.
Current address: Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America
Alaskan Husky Encephalopathy (AHE) has been previously proposed as a mitochondrial encephalopathy based on neuropathological similarities with human Leigh Syndrome (LS). We studied 11 Alaskan Husky dogs with AHE, but found no abnormalities in respiratory chain enzyme activities in muscle and liver, or mutations in mitochondrial or nuclear genes that cause LS in people. A genome wide association study was performed using eight of the affected dogs and 20 related but unaffected control AHs using the Illumina canine HD array.
Alaskan Husky Encephalopathy (AHE) is a fatal brain disease in young Alaskan Husky (AH) dogs, often affecting multiple dogs from the same litter
Dogs with AHE may have acute onset of clinical signs, or chronic progressive waxing and waning clinical history. Typically, they have multifocal central nervous system deficits including seizures, altered mentation, dysphagia, absent menace response, central blindness, hypermetria, proprioceptive positioning deficits, facial hypoalgesia, ataxia and tetraparesis.
Diagnostic testing reported in dogs with AHE was limited to normal serum and cerebrospinal fluid pyruvate and lactate concentrations; evaluation of mitochondrial respiratory chain enzymes was not done. Two dogs also had intracranial imaging. Computed tomography images in one dog had bilateral hypoattenuating lesions in the thalamus, and MR images of the other dog had bilateral hyperintense lesions in the brainstem on T2 weighted images. All dogs died or were euthanized, most within 2–7 months, however one dog lived for 1 year after the onset of clinical signs when it died of “natural causes”
In people and domestic animals, a multitude of uncommon diseases are associated with bilateral and symmetrically distributed brain lesions, apparent on magnetic resonance imaging and/or neuropathology. Examples include toxicities (carbon monoxide poisoning), metabolic (Leigh Syndrome
In people, mutations in the
All 11 dogs with AHE included in this study showed typical clinical signs (
MRI images (T2-weighted, transverse) of a normal brain from a clinically normal 1 year old male Alaskan Husky (C#1).
Dog # | Sex | Age at presentation (months) | Seizures | Tetraparesis | Generalized Ataxia | Thoracic Limb Hyper-metria | Dysphagia | Central Blindness | Longterm Outcome |
1 | F | 8 | Y | N | Y | Y | N | N | Euthanized |
2 | M | 8 | Y | Y | Y | Y | N | N | Euthanized 27 months post diagnosis |
3 | F | 12 | N | Y | Y | N | Y | Y | Euthanized |
4 | F | 6 | Y | Y | Y | Y | N | N | Alive |
5 | M | 6 | Y | Y | Y | Y | N | N | Euthanized 6 months post diagnosis |
6 | F | 8 | N | Y | Y | N | Y | Y | Euthanized |
7 | F | 21 | N | Y | Y | N | Y | N | Alive 27 months post diagnosis |
8 | M | 9 | Y | Y | Y | N | N | N | Euthanized |
9 | F | 24 | Y | N | N | N | N | N | Euthanized |
10 | M | 8 | N | N | Y | Y | Y | N | Alive 7 months post diagnosis |
11 | M | 6 | Y | Y | Y | N | N | N | Euthanized |
Dog #2 clinically improved and was neurologically stable 6 months after presentation. He had mild neurological deficits consisting of proprioceptive placing deficits in all four limbs, and infrequent generalized seizures. Clinical signs were static for the next 21 months, followed by a rapid progression. A repeat MRI showed increased size of the previously defined lesions. The dog was euthanized 27 months after presentation. Dog #4 is alive 51 months after initial presentation with continuing but static neurological deficits. An MRI was repeated 12 months after the initial MRI, and no changes were noted. Dog #5 had static neurological deficits for 5 months after presentation, but then developed progressive clinical signs, and was euthanized one month later. Repeat MRI 4 months after the initial one demonstrated increased size of the previously noted lesions. Dog #7 is alive with static neurological deficits 32 months after diagnosis; Dog #10 is alive 12 months after initial presentation. However seizures, worsening ataxia and hypermetria developed 4 months after being hit by a car and fracturing the pelvis. The neurological deficits have been static for the previous 9 months.
One normal AH dog with a normal physical, neurological examination and brain MRI evaluation was admitted into this study (C#1). Blood samples from 41 apparently healthy normal AH dogs from three racing kennels were obtained for genotyping.
Blood samples for genotyping were obtained from 187 randomly chosen canine patients from the VMTH including 51 breeds of both sexes and of varying ages, with a broad spectrum of clinical diseases. A skin biopsy sample for fibroblast culture was obtained from one dog with intervertebral disc disease at the time of laminectomy. Muscle biopsies were collected from one dog with a fractured femur (from the non-fracture leg) immediately after euthanasia.
Respiratory chain enzyme activities were measured in mitochondria from liver
Dog # | CI+III |
CI+III/CS | CII+III (nmol/mg/min) | CII+III/CS | CIV (nmol/mg/min) | CS (nmol/mg/min) | CIV/CS |
C#1 | 1.53 | 0.30 | 4.64 | 0.90 | 10.81 | 5.15 | 2.10 |
1 | 2.27 | 0.51 | 3.24 | 0.74 | 14.20 | 4.41 | 3.22 |
3 | 2.02 | 0.47 | 1.00 | 0.23 | 4.55 | 4.27 | 1.07 |
4 | 1.34 | 0.36 | 1.58 | 0.42 | 7.68 | 3.73 | 2.06 |
5 | 1.54 | 0.29 | 1.98 | 0.37 | 12.60 | 5.35 | 2.36 |
CI+III (complex I+III), NADH: cytochrome c reductase; CII+III (complex II+III), succinate:cytochrome c reductase; CIV (complex IV), cytochrome oxidase; CS (citrate synthase).
Complex | Dog #1 | Dog #2 | Dog #2 (repeat) | Control non AH Dog |
I+III | 1.05 | 1.01 | 3.45 | 0.72 |
I | 31.74 | 30.66 | 35.91 | 28.52 |
II+III | 0.74 | 0.73 | 0.91 | 0.32 |
IV | 2.66 | 2.65 | 3.23 | 2.30 |
SDH | 1.41 | 1.15 | 1.77 | 0.64 |
CS | 40.43 | 31.29 | 34.96 | 33.70 |
Values/CS | ||||
I+III | 0.026 | 0.032 | 0.098 | 0.021 |
I | 0.79 | 0.98 | 1.03 | 0.85 |
II+III | 0.018 | 0.023 | 0.026 | 0.0094 |
IV | 0.066 | 0.085 | 0.092 | 0.068 |
SDH | 0.035 | 0.037 | 0.051 | 0.019 |
Values/SDH | ||||
I+III | 0.74 | 0.88 | 1.95 | 1.12 |
I | 22.51 | 26.7 | 20.20 | 52.66 |
II+III | 0.52 | 0.63 | 0.51 | 0.50 |
IV | 1.89 | 2.30 | 1.82 | 3.59 |
CS | 28.67 | 27.2 | 19.75 | 52.7 |
SDH = succinate:dehydrogenase; CS = citrate synthase.
No abnormalities were noted on the CBC, CSF and serum biochemistry analysis in any dog. There was no indication of a mitochondrial disorder on evaluation of plasma or CSF lactate or pyruvate levels, or in the lactate to pyruvate ratios.
Seven dogs were euthanized and 5 had a necropsy. In 2 dogs, only the brain was evaluated pathologically. All dogs had neuropathological lesions characteristic for AHE with minor differences in severity among dogs, but not in location. In all dogs, bilaterally symmetrical areas of cystic encephalomalacia were grossly visible in the thalamus
Frozen fresh muscle was histochemically stained and evaluated by light microscopy in 2 dogs with AHE. In one dog (dog #1) the muscle was normal. In the other more severely affected dog (dog #2), deposits of succinic dehydrogenase (
There are areas of excessive SDH positive staining (arrow) in two myofibers, most consistent with mild mitochondrial proliferation.
Transmission electron microscopy. Glycogen deposits (arrowhead) are present in the abnormal megamitochondria (arrow).
Many of the mitochondrial DNA candidates for human Leigh Syndrome were sequenced in affected dogs (MT-ATP8/6, MT-tRNALYS, MT-tRNALEU(UUR), MT-tRNASER(AGY), MT-tRNATRY, MT-tRNATYR, MT-tRNAVAL) with no mutations identified in one of either blood, liver or fibroblast DNA.
In addition, the following genes were also sequenced:
Whole genome association was performed using 8 AHE affected dogs (dogs #1–8) and 20 unaffected but related controls. The genomic inflation factor for this group of samples was 1 indicating that there was little population stratification. After pruning, 114,613 SNPs were available for analysis. Raw p values showed a cluster of SNPs located on CFA (
A. Manhattan plot of –log 10 of raw p values (y axis) by chromosome (x axis). The best associated SNP (p = 6.59×10−6) was located at 43980115 Mb on Cfa 25 B. –log 10 of the permuted (100,000)p values (y axis) are plotted by chromosome (x axis). The lowest p value (0.051) was obtained for the same SNP located at 43980115 Mb. C. The Chi square value shown in grey and the allele frequency shown in black on the y axis are plotted against the Mb (x axis) on CFA 25. The location of the two paralogs of
Evaluation of this region for candidate genes uncovered SLC19A3 as a likely candidate, based on the phenotypic similarities between people affected with biotin-responsive basal ganglia disease (BBGD) and dogs with AHE. In the canine genome there has been a duplication of
RT PCR products obtained from equal amounts of cDNA are shown from the following tissues: 1 spleen, 2 skin, 3 cerebellum, 4 thymus, 5 testis, 6 spinal cord, 7 heart, 8 muscle, 9 cerebral cortex, 10 kidney, and 11 liver.
The coding region of
A. wildtype sequence, B. Heterozygous carrier, C. Mutant sequence.
RT PCR from equal amounts of RNA isolated from a normal beagle and an Alaskan husky sled dog affected with AHE.
Mutation | Location | Change in DNA (5′ to 3′) | Change in Protein |
12 bp insertion intron 1 | chr25:43,417,532 [191 bp downstream from ex2 (chr25:43,417,341)] |
|
none |
SNP 1 exon 2 | chr25:43,417,105 | C>T | none |
SNP 2 exon 2 + 4 bp insertion | chr25:43,416,868 | G>T + TTGC | Amino Acid 208: Gln>His, out of frame for 10 AA, protein termination at AA 219 |
8 bp deletion intron 2 | chr25:43,414,774-43,414,781 [79 bp downstream ex 3 (chr25:43,414,702)] | AAATAAAT | none |
SNP exon 5 | chr25:43,411,974 | A>G | Amino Acid 490: Thr>Ala |
We have demonstrated in this study that a homozygous mutation in the
Solute carrier family 19 (
Thiamine is an essential nutrient and thiamine deficiency (TD) may cause bilaterally symmetrical brain lesions in domestic animals and people
While dogs with TD and AHE have the same histological lesions of bilaterally symmetrical encephalomalacia, anatomic patterns of lesion distribution are distinctively different. Dogs with AHE do not have lesions in the caudal colliculus or lateral geniculate nucleus, and dogs with TD do not have thalamic lesions
There is a phenotypic spectrum of diseases in people associated with abnormalities in different regions of the
WLE was reported in 2 Japanese brothers in their second decade of life. Clinical signs included diplopia, external ophthalmoplegia, ptosis, seizures, nystagmus and ataxia. On brain MRI there were hyperintensities on the FLAIR (fluid attenuated inversion recovery) sequences bilaterally in the thalamus and periaqueductal region, which are entirely consistent with actual Wernicke's encephalopathy (thiamine deficiency encephalopathy). A compound heterozygous mutation of p.K44E and p.E320Q was found in
In four related Japanese boys with infantile seizures, psychomotor retardation, and characteristic lesions on brain MRI (focal T2W hyperintensity in bilateral symmetrical thalamic and basal nuclei, with cerebellar and cerebral cortical atrophy), a homozygous mutation was found (c.958G>C, p.E320Q) in
In BBGD, WLE, and in the Japanese boy encephalopathy, there is a distinct genotype-phenotype correlation. Patients affected by each disease have a remarkably similar phenotype. While the BBGD phenotype is thought to be secondary to a loss-of-function mutation, the Japanese boy encephalopathy may be due to a toxic gain-of-function secondary to the
Dogs with AHE did not have any biochemical evidence of primary mitochondrial disease in liver, skeletal muscle or fibroblasts, but did have subtle evidence of mitochondrial pathology. By light microscopy, there was prominent mitochondrial hyperplasia in muscle and liver and ultrastructural examination of muscle revealed “megamitochondria” and abnormal glycogen deposits. However such abnormal mitochondrial morphology is not necessarily evidence of a primary mitochondrial disease. The observed mitochondrial changes could be epiphenomenal, reflective of a response to a defect in energy metabolic pathways as expected with a defective
In conclusion, we have demonstrated that a homozygous mutation in
Alaskan Husky dogs with clinical signs suggestive of AHE, including generalized seizures, gait abnormalities, dysphagia and ataxia
Inclusion criteria for the one healthy, negative control AH was an age greater than one year of age, normal physical and neurological examination, and normal MRI of the brain with no detectable lesions in T1-weighted (T1W) pre- and post contrast, T2-weighted (T2W), proton density weighted, and FLAIR images in both the transverse and sagittal planes. Inclusion criteria for the other dogs were that they were reportedly healthy, older than one year of age, actively racing AHs of both sexes.
Non-Alaskan Husky control dogs were randomly selected from canine patients of both sexes and of varying ages admitted to the VMTH during the study period. These control dogs had a wide range of clinical diseases such as kidney disease and intervertebral disc disease, but none had similar clinical signs to AHE. All studies were done with their owner's consent, and in strict accordance with good animal practice, with study protocols approved by the Institutional Animal Care and Use Committee (IACUC) at UC Davis.
Skin biopsies were obtained and fibroblasts were grown and maintained in Eagle's minimal essential medium culture medium, containing 2 mM glutamine (Invitrogen, Burlington, ON, Canada), 1% penicillin/streptomycin and 20% fetal calf serum (Wisent, St-Bruno, Quebec, Canada).
Skeletal muscle and liver specimens were collected under general anesthesia or at necropsy and immediately frozen. Enzyme activities in both skeletal muscle and liver samples were assayed for NADH:cytochrome
Complete blood cell counts, serum biochemistry profiles, urinalysis and CSF analysis were done at the VMTH. Pyruvate and lactate concentrations were measured in both CSF and plasma. Plasma samples were obtained at rest and following 10 minutes of strenuous exercise were collected and evaluated at the Comparative Neuromuscular Laboratory, School of Medicine, UC San Diego, as previously described
Fresh tissue samples (skin, skeletal muscle and liver) were obtained either by surgical biopsy under general anesthesia or immediately after euthanasia. Muscle biopsy specimens were collected from the vastus lateralis and triceps brachii muscles and frozen immediately using isopentane precooled in liquid nitrogen. A standard panel of histochemical stains and reactions was performed on cryostat sections (10 μm thick) including HE, myofibrillar adenosine triphosphatase at pH 9.8, 4.5 and 4.3, modified Gomori trichrome, periodic acid-Schiff hematoxylin, esterase, staphylococcal protein A-horseradish peroxidase, oil red O, nicotinamide adenine dinucleotide tetrazolium reductase,and acid phosphatase as previously described
Dogs with AHE euthanized at the VMTH had an immediate necropsy evaluation. The CNS and other issues were immersion-fixed in 10% formalin, and selected tissues were then routinely embedded in paraffin, sectioned at 5 μm and evaluated after HE staining. Selected brain sections were also histochemically stained with HE-Luxol fast blue. For transmission electron microscopy, skeletal muscle specimens were immersion fixed in 3% glutaraldehyde in buffer, routinely processed and embedded in plastic resin. Thick sections were stained with toluidine blue and appropriate thin sections selected for transmission electron microscopic evaluation as previously described
Genomic DNA was isolated from cultured fibroblasts, blood or tissue using a Puregene genomic DNA isolation kit (Inter Medico, Markham, ON, Canada). RNA was isolated using Trizol (Life Technologies, Burlington, ON, Canada). Full length cDNA sequences were generated using RT-PCR and Superscript II reverse transcriptase and amplified using Platinum Hi-Fi Taq polymerase (Life Technologies, Burlington, ON, Canada). Oligonucleotide primers used for routine PCR (IDT, Iowa, USA) are listed in
Forward primer | Reverse primer | |
mtDNA (gDNA) | ||
MT-tRNALYS, MT-ATP8, MT-ATP6 | DogATP8F |
DogATP6R |
MT-tRNAVAL | Dog tRNAVal-F |
Dog tRNAVal-R |
MT-tRNALEU(UUR) | DogtRNA-LeuF |
DogtRNA-LeuR |
MT-tRNASER(AGY) | DogtRNA-Ser2F |
DogtRNA-Ser2R |
MT-tRNATRY, MT-tRNATYR | DogtRNA-TrpF |
DogtRNA-TyrR |
Nuclear genes (cDNA unless noted) | ||
DogPOLG-F |
POLG-R5 |
|
DogPOLG-F1 |
DogPOLG-R1 |
|
DogPOLG-F2 |
DogPOLG-R2 |
|
DogPOLG-F3 |
DogPOLG-R3 |
|
Twinkle | Dog-twinkle-F |
Dog-twinkle-R |
TK2 (in two parts) (gDNA) | Dog-TK2-F |
Dog-TK2-int1R |
(cDNA) | DogTK2-ex3F |
DogTK2-R2 |
DGUOK | DogDGUOK-F |
DogDGUOK-R |
SUCLG1 (2 parts) | DogSUCLG1-F |
DogSUCLG1-R1 |
DogSUCLG1-F1 |
DogSUCLG1-R |
|
SUCLA2 (in two parts) (gDNA) | Dog-SUCLA2-F |
Dog-SUCLA2-ex2R |
(cDNA) | Dog-SUCLA2-F3 |
Dog-SUCLA2-R1 |
MPV17 (3 parts) (gDNA) | DogMPV17-F4 |
DogMPV17int1R |
DogMPV17int1F |
DogMPV17int4R |
|
DogMPV17int4F |
DogMPV17int6R |
|
RRMB2 | Dog-RRM2B-F |
Dog-RRM2B-R |
SLC19A3.1 (gDNA) | Exon 1 |
|
Exon 2 |
|
|
Exon 3 |
|
|
Exon 4 5′-TGGTGTTTTAATTCAGCCTTCA-3 |
|
|
Exon 5 |
|
|
SLC19A3.2 |
|
|
SLC19A3 |
|
|
GAPDH |
|
|
DNA extracted from blood samples from AHE and control dogs was genotyped using the Illumina HD canine array (Illumina Inc, San Diego, California). Association analysis was performed using PLINK software
To verify the structure of exon 2 in paralog
Each of the 5 exons of
PCR was performed using primers that allowed amplification of all exons and intron/exon boundaries. Reactions were performed on an Applied Biosystems Gene Amp PCR System 9700. Each reaction consisted of 13.9 µl water, 2 µl 10X buffer with MgCl2, 1 µl dNTP, 1 µl of each primer, 0.1 µl of DNA Taq Polymerase, and 1 µl of DNA. Cycling programs were based on the primers' Tm and the expected product size. Amplified samples were then sequenced on an Applied Biosystems 3100 Genetic Analyzer using the Big Dye Terminator Sequencing Kit. Sequences were aligned using VectorNTI software (Life Technologies, Burlington, ON, Canada).
In order to screen a large number of normal control dogs, comprising both AH and other breed dogs, for the 4 bp insertion and SNP identified in exon 2, a PCR genotyping assay was developed using the following primers: 5′ 6FAM-ATCCTTGGCCTCTGTCTGTG,
Semiquantitive RT-PCR was performed to compare the tissue expression of
We wish to acknowledge the following individuals for their help in this study, from caring for these dogs with AHE, to the completion of this manuscript: John Doval, Rich Larson, Kathy Pinkston, Maggie Knipe, Daniel York, Neils Pedersen, Andrea Engwis, Hongwei Lui, Nicole Lombardi, Dawn, Terry and Jeanne Parmeter, Dr. Deidre Puaio, Dr. Catherine Colangelo, Daniel Ralph, Dr. Sarah Love, Dr. Joe Wakshlag, Lori Gildhaus, Jeffrey Wells, Nicole Torre, Carrie Skinner, and to Alaskan Huskies Dogs: McGruff, Tipsy, Walker, Nanuk, Denali and Rubl.