Abstract
Polyglucosan bodies are accumulations of insoluble glucose polymers and proteins that form intracytoplasmic inclusions in the brain, large numbers of which can be indicative of neurodegenerative diseases such as Lafora disease. Montserrat orioles (Icterus oberi) are an icterid passerine endemic to Montserrat with conservation populations maintained in captivity abroad. We demonstrate that polyglucosan bodies are unusually abundant in the cerebellar molecular and Purkinje cell layers and cerebellar peduncles of captive-bred and wild-caught Montserrat orioles. The bodies are periodic acid-Schiff positive and diastase resistant and label with concanavalin A and for ubiquitin, consistent with those seen in humans. We found no association of the polyglucosan bodies with concurrent neurological lesions or clinical signs, nor with EPM2A and EPM2B gene mutations associated with Lafora disease. We conclude that an abundance of cerebellar polyglucosan bodies may be a normal finding in aged Montserrat orioles and not a threat to the captive breeding population.
Polyglucosan bodies (PGBs) are nonmembrane bound intracytoplasmic inclusion bodies comprising insoluble glucose polymers 15 found predominantly in the brain, where they may be neuronal, glial, or extracellular, and in cardiac myofibers. Polyglucosan bodies are primarily composed of amylopectin-like, branched glucosans 5 with a smaller fraction (4%) of assorted proteins, many of which were ubiquitinated, 4 and appear as round to oval, basophilic, periodic acid-Schiff positive, diastase resistant, 2 to 20 µm globular inclusions. The prevalence of PGBs increases with age in humans and many other mammals in the absence of associated disease; these PGBs are usually termed “corpora amylacea” (“amyloid body,” a misnomer as they do not typically contain amyloid) and are considered a normal feature of aging. 10 However, copious numbers of PGBs are key features of several severe, inherited human neurological diseases, including Lafora disease, caused by EPM2A or EPM2B mutations, 8 and adult PGB disease, caused by GBE1 mutations. 9 Lafora-like diseases have been reported in several animal species, including cows, cats, fennec foxes, and parrots, but the genetic cause has only been determined in dogs, where there is a repeat expansion in EPM2B.3,8,16 As these diseases show autosomal recessive inheritance, the presence of defective alleles is particularly detrimental in small populations where inbreeding is likely, such as endangered species in captive breeding programs.
Montserrat orioles (Icterus oberi), hereon referred to as orioles, are icterid passerines endemic to the Caribbean island of Montserrat. 1 Owing to habitat loss associated with volcanic activity and invasive species, the species was listed as critically endangered. 1 Toward breeding and conservation goals, a captive population was established at Jersey Zoo (Durrell Wildlife Conservation Trust) in the 1990s.6,11 Wild populations have since stabilized, but the species is currently classified as vulnerable by the International Union for Conservation of Nature Red List. 1 Estimates predict only 250 to 460 mature individuals remain in the wild, 1 so it is still crucial to maintain captive populations in good health.
In June of 2020, a postmortem examination was performed on an aged oriole (case 1) from a population held at the Zoological Society of London’s London Zoo, which had been euthanized due to age-related, degenerative osteoarthritis. While there was no clinical history of neurological disease, routine histological examination with hematoxylin and eosin unexpectedly revealed the presence of copious, 2 to 10 µm diameter, round to elongated, lightly basophilic, granular inclusions in the neuropil of the brain (Fig. 1a), but not in any other organ examined (spinal cord was not available). Further histochemical investigation revealed that the inclusions stained bright magenta with periodic acid-Schiff, which was not diminished by preceding diastase treatment. Electron microscopy showed a faintly targetoid, nonmembrane bound body with a granular core and a homogenous outer layer (Fig. 1b). Inclusions were found randomly, in low numbers, throughout the brain but were most densely clustered in the molecular and Purkinje cell layers of the cerebellum, where they formed linear arrays perpendicular to the layers (Fig. 1c), and in the cerebellar peduncle (Fig. 1d). There were no other findings in the brain.

Histological and ultrastructural presentation of polyglucosan bodies in the cerebellum of Montserrat orioles; all images are of case 1. (a) The bodies are oval, 2-10 µm long, and stain lightly basophilic with hematoxylin and eosin. Hematoxylin and eosin. (b) Ultrastructurally, the bodies (*) are intracytoplasmic, nonmembrane bound, and have targetoid density, with a granular center and a homogenous outer. The angular spaces may represent fixation artifact. Arrows, nuclei; closed arrowheads, mitochondria; open arrowheads, myelin figures. Bar is 2 µm. (c) In the molecular layer, the bodies are arranged in linear arrays, suggesting an axonal distribution. Periodic acid-Schiff. (d) In the peduncle, the bodies are round and randomly arranged, sometimes within the perikarya of neurons and otherwise free in the neuropil. Periodic acid-Schiff.
The inclusions resembled PGBs but were present in very high numbers. There were 979 distinct periodic acid-Schiff-stained inclusions in a single 0.1 mm2 area of cerebellar molecular layer, as quantified by intensity thresholding and the “analyze particles” function of ImageJ in the green channel. This was likely an underestimate as many of the larger bodies were coalescing and so undercounted. While the presence of PGBs appeared incidental in this case, there was concern that this phenotype could be the consequence of heterozygous carriage of a mutant allele, as is seen in glycogen storage disease type 2 in humans, where carriers of the mutated GAA allele experience a mild, late onset version of the disease. 18 Such a lesion had not been previously reported in orioles or any other passerine. As the captive oriole population all descend from a founder population of just eight wild-caught individuals, 11 we sought to investigate the incidence of PGBs in the captive population, to determine if they were associated with any clinical signs, neurological lesions, or reproductive failure and to further characterize their composition and characteristics.
To survey the prevalence of PGBs in the captive population, brain tissue was collected from the archives of the Zoological Society of London, Durrell Wildlife Conservation Trust, and the International Zoo Veterinary Group. Nineteen oriole brains with cerebellum were available for examination (Supplemental Table S1). The age at death ranged from 14 days to 20 years and 11 months, with a median age of 3 years. There were 8 males, 6 females, and 5 animals of undetermined sex. Brains were scored by the presence of segmental clusters containing low (5-50), medium (50-200), or high (>200) numbers of PGBs, or as negative if only present in clusters of <5. Polyglucosan bodies were seen in 8 animals (Table 1); of the 6 animals over 8 years old, all were positive, and the youngest positive animal was 2 years and 8 months old, with low numbers of PGBs. Five of the positive cases were male and 3 were female. None of the cases had histories of neurological disease, all had definitively identified causes of death, and no additional neurological lesions were identified histologically.
Age, sex, cause of death, genetic background, and abundance of polyglucosan bodies (PGBs) in the brains of Montserrat orioles.
Abbreviations: F, female; M, male; m, months; y, years.
The level of PGBs is defined as: low, segmental clusters of 5-50 PGBs; medium, 50-200; and high, >200. Clusters of <5 PGBs were considered incidental and scored as negative.
The eight wild-caught founder animals are assigned codes A-H; genetic background is then given as the codes of the founders from which each case is descended.
To further characterize the bodies, lectin- and immunohistochemistry were performed to gain information on the carbohydrate sequences and assess the expression of neurodegenerative-associated proteins, respectively (Supplemental Methods). The bodies strongly labeled by concanavalin A (Fig. 2a) and for ubiquitin (Fig. 2b), but not the b-galactose-specific peanut agglutinin, the N-acetyl-

Lectin histochemistry and immunohistochemistry of polyglucosan bodies in the cerebellum of Montserrat orioles. (a) The bodies label with concanavalin A, which is inhibited by 0.5 M monosaccharide (inset). Case 1. Concanavalin A lectin histochemistry. (b) The bodies label strongly for ubiquitin; the secondary antibody alone does not label (inset). Case 5. Ubiquitin immunohistochemistry.
Using records from the Zoological Information Management System (Species360, zims. species360.org), the relationships of the cases were deduced (Table 1). Case 4 was a wild-caught animal and cases 1, 2, and 7 were first generation offspring of two pairings of four other founders that were not available for this study. This suggests that if there is a genetic basis to the PGBs, it must be present in at least three of the eight founder individuals. As these animals were wild caught, their consanguinity is unknown, but these results suggest that predisposition to PGBs is common in the wild population. Lafora disease was considered unlikely in the cases examined due to the lack of clinical disease, the absence of typical radial filament appearance of the PGBs, and the lack of PGBs in other organs. However, the concern remained that these animals may be heterozygotes for the condition. Human and canine Lafora disease is caused by mutations in either EPM2A or EPM2B, which respectively code for laforin, a carbohydrate phosphatase, and malin, a ubiquitin ligase, and are hypothesized to result in the precipitation of insoluble glycogen derivates. 7 To rule out the presence of such mutations, EPM2A and EPM2B were amplified from frozen skeletal muscle from case 1, and frozen muscle or liver from two unaffected controls (cases 9 and 10, Supplemental Table S1), and sequenced (Supplemental Methods). The coding sequence of EPM2A or EPM2B showed no insertions, deletions, or premature stop codons, and there were no missense mutations in case 1 relative to the unaffected controls. The coding sequences of the two candidate genes were therefore devoid of any putatively pathological mutations that could contribute to a Lafora-like phenotype.
Age was the most important correlate for high levels of PGBs in orioles, with them present in all animals over eight years of age, and at high levels in all animals over 19. The maximum lifespan of wild Montserrat orioles is unknown, but in other Icterus species it ranges from 6 to 14 years. 17 Thus, animals over 19 are likely senescent. It has recently been hypothesized that PGBs are transient organelles formed of a glucosan skeleton filled with ubiquitin-labeled cell waste, termed “wasteosomes.” 14 Wasteosomes are normally excreted from cells and phagocytosed by macrophages for removal from the tissue. 13 Thus, the unusually high levels of PGBs in the brains of older orioles may represent reduced ability to excrete wasteosomes causing them to accumulate in axons, albeit harmlessly. However, wasteosomes can also contain exogenous waste, including DNA and protein of microorganisms. 12 Two of the eight positive animals in the study died from yersiniosis but it is not possible to say whether the infections were in any way affected by decreased wasteosome clearance, vice versa, or whether the findings are completely independent.
In summary, PGBs are unusually abundant in the cerebellum of aged orioles, both in wild-caught and captive-bred animals. The composition and histological appearance of the bodies resembles that of those seen in aged humans, 2 there are no associated neurological lesions or clinical disease, and, in case 1, there is no evidence of association with EPM2A or EPM2B mutations. We conclude that PGBs are a normal, age-related finding in Montserrat orioles and do not pose a threat to the captive breeding of this vulnerable species.
Supplemental Material
sj-pdf-1-vet-10.1177_03009858241270000 – Supplemental material for Incidence and characterization of polyglucosan bodies in the cerebella of montserrat orioles (Icterus oberi)
Supplemental material, sj-pdf-1-vet-10.1177_03009858241270000 for Incidence and characterization of polyglucosan bodies in the cerebella of montserrat orioles (Icterus oberi) by Simon Spiro, Marta Pereira, Kieran A. Bates, Zane Jaunmuktane, David J. Everest, Mark F. Stidworthy, Daniela Denk, Alejandro Núñez, Ethan Wrigglesworth, Andrew Theodoulou, Alberto Barbon, Emma Nye, Yan Liu, Adrian L. Smith and Steven Fiddaman in Veterinary Pathology
Footnotes
Acknowledgements
The authors thank the government of Montserrat, who are the owners of all the orioles in this study, for allowing them to study their animals. They thank Lorraine Lawrence for assistance with histology.
Authors’ Contributions
SS, MP, and SF conceived and designed the study; SS, EW, MFS, DD, and AB performed gross postmortem examinations; SS, MP, MFS, DD, and ZJ performed histological examinations; DJE and AN performed ultrastructural examinations; KAB and YL performed lectin histochemistry; ZJ and AT performed immunohistochemistry; EW and EN provided technical support; ALS and SF performed genetic investigations; the manuscript was written by SS and MP with contributions from all authors.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Author MP was funded by a grant from the Zebra Foundation for Zoological Education and by the MSc Wild Animal Health programme of the Royal Veterinary College.
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
