- Top of page
- Materials and Methods
Background: A missense mutation in the GYS1 gene was recently described in horses with polysaccharide storage myopathy (PSSM).
Objectives: The first objective was to determine the prevalence of the GYS1 mutation in horses with PSSM from diverse breeds. The second objective was to determine if the prevalence of the GYS1 mutation differed between horses diagnosed with PSSM based on grade 1 (typically amylase-sensitive) or grade 2 (typically amylase-resistant) polysaccharide.
Animals: Eight hundred and thirty-one PSSM horses from 36 breeds.
Procedures: Horses with PSSM diagnosed by histopathology of skeletal muscle biopsy samples were identified from the Neuromuscular Disease Laboratory database. Eight hundred and thirty-one cases had blood or tissue that was available for DNA isolation; these 831 cases were genotyped for the GYS1 mutation by restriction fragment length polymorphism.
Results: The PSSM mutation was identified in horses from 17 different breeds. The prevalence of the GYS1 mutation in PSSM horses was high in Draft- (87%) and Quarter Horse-related breeds (72%) and lower in Warmbloods (18%) and other light horse breeds (24%), when diagnosis was based on grade 2 diagnostic criteria. Overall, the PSSM mutation was present in 16% of grade 1 and 70% of grade 2 PSSM horses.
Conclusions and Clinical Importance: GYS1 mutation causes PSSM in diverse breeds and is the predominant form of PSSM in Draft- and Quarter Horse-related breeds. False-positive diagnosis, as well as the possibility of a second glycogenosis in horses with neuromuscular disease (type 2 PSSM), might explain the absence of the GYS1 mutation in horses diagnosed with excessive glycogen accumulation in muscle.
A novel glycogen storage disease termed polysaccharide storage myopathy (PSSM) was originally reported in 9 horses with chronic exertional rhabdomyolysis (Quarter Horses and Quarter Horse crosses, an American Paint Horse and 2 Appaloosas).1 In this original study, histopathologic features of PSSM included the presence of subsarcolemmal vacuoles increased periodic acid Schiff's (PAS)-positive staining for muscle glycogen and, in each case, the presence of PAS positive amylase-resistant abnormal crystalline polysaccharide inclusion.1 The common causes of glycogen storage diseases in other species, such as enzymatic deficiencies in glycogenolysis, glycolysis, lysosomal degradation, or AMP-activated protein kinase (AMPK),2,3 were not present in these PSSM horses.4,5
Belgian and Percheron Draft Horses with rhabdomyolysis were subsequently diagnosed with PSSM based on the finding of amylase-resistant abnormal polysaccharide in skeletal muscle fibers.6 Similar to Quarter Horses, glycogenolytic and glycolytic enzyme activities were normal in several of these horses.7 It was not clear whether PSSM in these diverse horse breeds had the same metabolic or genetic basis.
A number of other breeds of horses were later diagnosed with PSSM on the basis of the original criteria of amylase-resistant polysaccharide as well as alternative diagnostic criteria.8,9 When the alternate criteria of excessive amylase-sensitive granular cytoplasmic or subsarcolemmal glycogen, or PAS-positive sarcoplasmic masses were used, rather than amylase-resistant PAS-positive inclusions, light breeds such as Arabian, Thoroughbred, Morgan, Welsh pony, and Standardbred, among others, were thought to have PSSM.6,8–10 A retrospective study of over 1,400 muscle biopsy specimens from horses with neuromuscular disease found that use of alternate histopathologic diagnostic criteria resulted in an increase in the number of muscle biopsies diagnosed with PSSM from 22 to 40%.11 Thus there was little agreement as to the best method to diagnose PSSM. For clarity, a grading system was developed in which muscle biopsies were graded as mild or grade 1 if they possess increased amylase-sensitive glycogen (Fig 1A) and moderate or grade 2 if they possess PAS-positive inclusion that are typically amylase resistant (Fig 1D).11
Figure 1. Periodic acid Schiff's (PAS) stains of skeletal muscle biopsy samples from polysaccharide storage myopathy horses (× 40). (A) Grade 1: aggregates of granular polysaccharide in the cytoplasm or under the sarcolemma (arrows). (B) Serial section of (A); aggregates are sensitive to amylase digestion (arrows). (C) Grade 2: PAS-positive, abnormal crystalline inclusions (arrows). (D) Crystalline inclusions are resistant to amylase digestion (arrows).
Download figure to PowerPoint
More sensitive and specific testing for PSSM awaited identification of a molecular defect. We have recently identified a mutation in the GYS1 gene of Quarter Horses with amylase-resistant polysaccharide accumulation in skeletal muscle. The GYS1 gene encodes the skeletal muscle isoform of glycogen synthase (GS) and a single base pair substitution that causes an amino acid change (Arg309His) in a highly conserved, potentially critical region of the GS enzyme was identified.12 The inheritance pattern and genotype distribution of this mutation in phenotyped horses fit those expected of a highly penetrant autosomal dominant trait. Further analyses of biopsy samples indicated that the same GYS1 mutation was also highly associated with PSSM in several other breeds.12 Haplotype analysis revealed that this mutation was identical by descent among the horse breeds studied, suggesting that the mutation, and resultant genetic susceptibility to PSSM, arose in a single horse long before the formation of the diverse horse breeds recognized today.12
The first objective of this study was to determine the prevalence of the GYS1 mutation in all breeds in which PSSM had been identified based on muscle biopsy specimens' submissions to the Neuromuscular Diagnostic Laboratory. The second objective was to determine whether the frequency of the GYS1 mutation differed between horses diagnosed with PSSM based on grades 1 or 2 diagnostic criteria.
- Top of page
- Materials and Methods
Overall the GYS1 H allele (the PSSM mutation) was present in 48.3% (401/831) of the PSSM cases genotyped. The allele frequencies of the R and H allele in all PSSM-affected horses utilized in this study were 0.65 and 0.35, respectively. Thirty-one homozygous affected (H/H), 370 heterozygous (R/H), and 430 homozygous normal horses were identified.
The H allele of the GYS1 gene was identified in PSSM horses from 17 breeds including Quarter Horses, Paint Horses, Appaloosa Horses, 5 Draft Horse breeds, 3 Warmblood breeds, Haflinger, Morgan, Mustang, Rocky Mountain Horse, and Tennessee Walking Horse breeds, as well as mixed breed horses and Warmblood Horses of unspecified breed (Tables 2 and 3). Within the mixed breed category, all of the PSSM horses with the GYS1 H allele that had breed information available were crosses of Quarter Horses, Draft, or Warmblood Horses. The GYS1 mutation was not identified in any of the Standardbred, Thoroughbred, or Arabian Horses diagnosed with PSSM (Tables 2 and 3).
Table 2. GYS1 genotyping results for all grade 1 PSSM cases in each breed and breed category.
|Quarter Horse||2 (1.2)||38 (22.8)||127 (76.0)||167||24.0|
|Appendix QH||0||0||1 (100)||1||0.0|
|Appaloosa||0||1 (9.1)||10 (90.9)||11||9.1|
|Paint||0||5 (18.5)||22 (81.5)||27||18.5|
|Quarter Horse-related breeds||2 (1.0)||44 (21.4)||160 (77.7)||206||22.3|
|Belgian||0||2 (18.2)||9 (81.8)||11||18.2|
|Irish Draft||0||0||1 (100)||1||0.0|
|Percheron||1 (14.3)||1 (14.3)||5 (71.4)||7||28.6|
|Draft breeds||1 (3.7)||3 (11.11)||23 (85.2)||27||1.7|
|Dutch Warmblood||0||0||2 (100)||2||0.0|
|Warmblood breeds||0||1 (2.3)||43 (97.7)||44||2.3|
|Icelandic Horse||0||0||1 (100)||1||0.0|
|Morgan||0||1 (25)||3 (75)||4||25.0|
|Paso Fino||0||0 (0)||1 (100)||1||0.0|
|Rocky Mountain Horse||0||1 (33.3)||2 (66.7)||3||33.3|
|Tennessee Walking Horse||0||0||2 (100)||2||0.0|
|Welsh Cob||0||0||1 (100)||1||0.0|
|Other breeds||0||2 (5.0)||38 (95.0)||40||5.0|
|Total||3 (0.9)||51 (15.3)||282 (84.7)||336||16.1|
Table 3. GYS1 genotyping results for all grade 2 PSSM cases in each breed and breed category.
|Quarter Horse||10 (4.4)||157 (68.6)||62 (27.1)||229||72.9|
|Appendix QH||0||4 (44.4)||5 (55.6)||9||44.4|
|Appaloosa||2 (14.3)||10 (71.4)||2 (14.3)||14||85.7|
|Paint||1 (2.5)||26 (65)||13 (32.5)||40||67.5|
|Quarter Horse-related breeds||13 (4.5)||197 (67.5)||82 (28.1)||292||71.9|
|American Cream Draft||0||1 (100)||0||1||100.0|
|Belgian||8 (11.4)||54 (77.1)||8 (11.4)||70||88.6|
|Haflinger||1 (14.3)||4 (57.1)||2 (28.6)||7||71.4|
|Percheron||4 (21.1)||15 (78.9)||0||19||100.0|
|Suffolk Punch||0||1 (50)||1 (50)||2||50.0|
|Draft breeds||13 (12.7)||76 (74.5)||13 (12.7)||102||87.3|
|Dutch Warmblood||0||0||2 (100)||2||0.0|
|Hanoverian||0||1 (50)||1 (50)||2||50.0|
|Irish Sport Horse||0||0||1 (100)||1||0.0|
|Swedish Warmblood||0||0||2 (100)||2||0.0|
|Warmblood||0||4 (19.0)||17 (80.9)||21||19.0|
|Warmblood breeds||0||5 (17.9)||23 (82.1)||28||17.9|
|Mixed breed||2 (5.3)||27 (71.1)||9 (23.7)||38||76.3|
|Morgan||0||1 (50)||1 (50)||2||50.0|
|Mustang||0||3 (75)||1 (25)||4||75.0|
|Peruvian Paso||0||0||1 (100)||1||0.0|
|Tennessee Walking Horse||0||2 (66.7)||1 (33.3)||3||66.7|
|Thoroughbred||0||0 (0)||9 (100)||9||0.0|
|Other breeds||0||6 (24.0)||19 (76.0)||25||24.0|
|Unknown||0||8 (80.0)||2 (20.0)||10||80.0|
|Total||28 (5.7)||319 (64.4)||148 (29.9)||358||70.1|
Grade 1 PSSM horses comprised 336 of 831 (40.4%) samples and Grade 2 comprised 495 of the 831 horses (59.6%). The H allele was present in 16% of grade 1 (Table 2) and 70% of grade 2 (Table 3) PSSM cases. Among grade 1 cases, about 1% were homozygous affected (H/H), 14% heterozygous affected (R/H), and 85% were homozygous normal. Among grade 2 cases, 6% were homozygous affected, 64% were heterozygous affected, and 30% were homozygous normal cases.
- Top of page
- Materials and Methods
This study clearly demonstrates that the GYS1 mutation (H allele) is associated with PSSM in 17 of 36 horse breeds in the present study. The prevalence of the GYS1 mutation was highly variable in the breeds in which it was identified. Prevalence was as low as approximately 8% in Warmbloods with PSSM to as high as approximately 80% of Belgian and Percheron Draft crosses with PSSM (grades 1 and 2 combined). Thus, the GYS1 mutation is apparently an important, but not the only, cause of PSSM diagnosed by muscle biopsy.
Approximately half of Quarter Horses and related breeds diagnosed with PSSM by muscle biopsy (grades 1 and 2 combined) had the mutation in at least 1 copy of their GYS 1 gene. The similar prevalence of the GYS1 mutation in Quarter, American Paint, and Appaloosa Horses with PSSM was not surprising. PSSM horses in these breeds share many phenotypic characteristics with Quarter Horses,1,14 and both of these populations diverged from the Quarter Horse and they share multiple founders.15
PSSM was highly prevalent in Draft breeds in the present study. The GYS1 mutation was identified in 6 of 9 Draft breeds with PSSM, and affected 87% of the horses with a grade 2 muscle biopsy diagnosis. Further, the GYS1 H allele frequency was high in these populations, resulting in a high proportion of PSSM Belgian and Percheron Horses that were homozygous for the PSSM mutation (11%). The high prevalence of PSSM in these breeds may be related to the genetic bottleneck that occurred in North America after the Second World War and the resulting high degree of inbreeding.16 It is likely that some of the breeding animals that formed the foundation for many of the modern day North American Draft Horses were carriers of the GYS1 mutation, resulting in it now being an important cause of PSSM in Draft Horses and their crosses.
Although PSSM was diagnosed by muscle biopsy in 72 Warmbloods, only 6 horses, encompassing 3 of the 11 Warmblood breeds sampled, had the GYS1 mutation. When only grade 2 PSSM cases were considered, only 18% of Warmblood Horses with PSSM in this database had the GYS1 mutation. The results suggest that the remaining horses diagnosed with PSSM either have another glycogenosis and/or there may be a high proportion of false-positive muscle biopsy diagnoses, particularly in the grade 1 category. It was not possible to determine the distribution of the GYS1 mutation among specific Warmblood breeds because of the small sample sizes. Failure to detect the GYS1 H allele in these horses does not rule out the possibility that it plays a role in PSSM in these breeds. To determine the true prevalence of the GYS1 mutation in these breeds, a larger random sampling of individuals would be required to estimate the population frequency of the H and R alleles.
PSSM was also diagnosed in 19 other horses of a variety of light breeds. However, only Morgan, Mustang, Rocky Mountain, and Tennessee Walking Horses had the GYS1 H allele. Again, failure to detect the GYS1 H allele in a small sample size does not rule out the possibility that it plays a role in PSSM in some of these breeds. Notably the largest group tested was Thoroughbreds (34 horses) and none possessed the GYS1 mutation.
Previous studies have suggested that use of grade 2 versus grade 1 diagnostic criteria increases the specificity of a diagnosis of PSSM.11,14 In the present study, the proportion of grade 1 PSSM horses with the GYS1 mutation was 24% compared with 73% in grade 2 cases within Quarter Horse-related breeds. Similarly, the proportion of PSSM horses with the GYS1 mutation was 68 and 86% in grade 2 American Paint and Appaloosa Horses, respectively, compared with 9 and 19% in grade 1 cases, respectively. A dramatic difference (2 and 87%) in the proportion of PSSM horses with the GYS1 mutation was also seen in Draft Horses when grade 1 and grade 2 were compared. Thus, grade 2 criteria do appear to be most specifically associated with the GYS1 mutation. One explanation for the high association between grade 2 polysaccharide accumulation (typically amylase resistant) with the GYS1 mutation may be that GS activity in horses with the mutation is unregulated. Unregulated, higher GS activity relative to branching enzyme activity in GYS1 PSSM muscle may result in polysaccharide with more straight chains and a less highly branched structure, making it more resistant to amylase digestion than normal glycogen.17
The low prevalence of the GYS1 mutation in grade 1 cases may be attributed to a higher number of false-positive diagnoses of PSSM when a subjective assessment of polysaccharide accumulation is made. The prevalence of PSSM was estimated to be as high as 66–80% within Draft Horse-related breeds and 33% of all light horse breeds,8,18 when formalin-fixed tissues were evaluated and grade 1 criteria, including the PAS-positive sarcoplasmic masses, were used as diagnostic criteria. In contrast, when grade 2 criteria of amylase-resistant abnormal polysaccharide was applied to biopsy specimens of >100 Belgian Draft Horses and >150 Quarter Horses, the disease prevalence was estimated at 34 and 10%, respectively, less than half the estimation in previous studies.19,20 Thus, the subjective assessment of increased glycogen content in biopsy specimens may lead to an increase in false-positive diagnoses and an overestimation of the prevalence of PSSM.
Most skeletal muscle glycogenoses in humans, however, result in an accumulation of excess amylase-sensitive glycogen (grade 1) in skeletal muscle without accumulation of abnormal amylase-resistant polysaccharide (grade 2). In fact, glycogen branching enzyme deficiency, phosphofructokinase enzyme deficiency,21 and mutations in the γ 2 and 3 subunits of AMPK2,22 were the only previously known glycogenoses to cause abnormal polysaccharide accumulation until the discovery of the GYS1 mutation. The high proportion of grade 1 horses with signs of exertional rhabdomyolysis that did not possess the GYS1 mutation, as well as approximately 20% of Quarter Horse-related breeds with grade 2 PSSM that did not have the mutation in GYS1, are not all likely to be attributable to false-positive biopsy diagnoses. It is highly likely therefore that a proportion of PSSM horses have a second non-GYS1 glycogenosis that is not consistently characterized by amylase-resistant polysaccharide inclusions. Further biochemical, molecular, and genetic studies of this group of horses are necessary to determine if a second heritable form of PSSM exists in such horses. Identification of the molecular basis for this second glycogenosis would then subsequently aid in differentiating false-positive assessment of PAS staining intensity from a specific genetic defect. To make a distinction between GYS1 and non-GYS1 PSSM horses, we propose the use of a new nomenclature where type 1 PSSM designates horses with the GYS1 mutation and type 2 PSSM designates horses with non-GYS1 form(s) of PSSM. Use of this system mirrors the nomenclature used in human glycogenoses where the genetic diseases are numbered by the order of discovery of the mutation.
In conclusion, the GYS1 mutation was found in approximately 50% of all biopsy samples diagnosed with PSSM. This mutation causes PSSM in at least 17 different horse breeds and is the predominant form of PSSM in Draft- and Quarter Horse-related breeds. Muscle biopsies from horses with the GYS1 mutation are most often characterized by the presence of abnormal PAS-positive crystalline inclusions. False-positive diagnosis, as well as the likely possibility of a second glycogenosis in horses, may explain the absence of the GYS1 mutation in horses diagnosed with PSSM based only on excessive glycogen accumulation in muscle. At present, the most accurate means to screen for PSSM in Draft- and Quarter Horse-related breeds maybe to screen hair root samples for the GYS1 mutation followed by a muscle biopsy, if this test is negative for the mutation. In breeds with a low prevalence of the GYS1 mutation, such as Warmbloods, muscle biopsy and simultaneous testing for the GYS1 mutation may be the most expedient means at present to diagnose a glycogenosis.