Histochemical characteristics and distribution of lipofuscin and polyglucosan bodies in the brain of dogs more than 10 years old

The ageing process is accompanied by numerous changes in the brain of dogs, such as accumulation of amyloid, fibrosis of blood vessel walls and meninges, accumulation of lipofuscin, and the presence of polyglucosan bodies (PGBs), satellitosis and neuronophagia. In this study, the presence of lipofuscin and PGBs in various parts of the brain in dogs of different sexes and ages was examined. For this purpose, brain samples were stained using haematoxylin eosin, modified Ziehl Neelsen and Periodic acid Schiff (PAS) methods. Lipofuscin was visualised by Ziehl Neelsen and PAS methods of specific staining on the same brain tissue segments. Lipofuscin had accumulated in 93% of old (more than 10 years old) dog brains, mostly in neurons of the medulla oblongata. The percentage of age-related lipofuscin pigment in other examined brain tissue segments was lower than in the medulla oblongata. There was a small difference in the percentage of lipofuscin-positive individuals between the two staining methods. The presence of PGBs was established by the PAS method for the vast majority (about 93%) of the old dogs (more than 10 years old), while PGBs were not detected in the group of young dogs (up to 5 years old). However, PGBs occurred in all examined segments of the dog?s brain tissues (for each of the tissue types, from 90% to 93% of dogs were positive for PGBs). The results obtained the oldest dogs (15 years old) harboured PGBs both extracellularly and intracellularly, while in other dogs, only extracellular PGBs were seen. Lipofuscin was accumulated mostly in large neurons of olivary nuclei of the medulla oblongata. PGBs were confirmed in all examined segments of the brain tissue of dogs more than 10 years old. This is one of the numerous indications that old dogs could be a very good animal model for studying the normal ageing process or neurodegenerative diseases.


INTRODUCTION
The ageing process is a phenomenon that has been studied for centuries, but even today there is no complete and unique definition of age, nor a comprehensive explanation of the ageing process. During the ageing process, amyloid accumulates as extracellular deposits in the dog brain, while lipofuscin granules can be observed intracellularly in the neurons of dog brain (Torp et al., 2000, Rofina et al., 2003, Papaioannou, 2014, Nešić et al., 2017.
Lipofuscin accumulates slowly over time, more intensely in the final stages of the ageing process (Jung et al., 2010). Lipofuscin is an ageing pigment that accumulates in greater amounts in non-dividing cells (muscle cells, heart muscle cells and neurons) (Porta, 2002, Gray and Woulfe, 2005, Gilissen et al., 2016. Intracellular lipofuscin production is a very complex process, as biochemical reactions affecting production occur in lysosomes, cytoplasm and mitochondria (Jung et al., 2010). Lipofuscin accumulates due to incomplete degradation of proteins, lipids and damaged organelles. The accumulation of metabolic products and damaged intracellular components is the result of insufficient and reduced autophagy in postmitotic cells (Kurz et al., 2007, Kurz et al., 2010. Factors that promote oxidative processes in the cell increase accumulation of metabolic products, leading to an increased amount of lipofuscin (Brunk and Terman, 2002a). Lipofuscin can occupy a major portion (40 to 75%) of the cell volume. Such a large amount of pigment fills the cytoplasm and initially results in a decrease of functional capacity, but later leads to apoptotic cell death (Jung et al., 2007).
Polyglucosan bodies (PGBs) occur in the brain and other organs of humans and animals. There are several forms of PGBs, such as corpora amylacea (CA), Lafora bodies, Lafora-like bodies, and others (Gredal et al., 2003, Márquez et al., 2010. CA in humans are analogous with PGBs in dogs, and they occur physiologically in humans during the ageing process. Lafora bodies are intracellular bodies found in patients with myoclonic epilepsy. PGBs, both intracellular and extracellular, occur normally in the ageing process of dog and cat brains, and are designated as Lafora-like bodies (Cavanagh, 1999, Borras et al., 1999, Lohi et al., 2005.
The aim of this study was, for the first time, to prove the presence and distribution of lipofuscin and PGBs in the brain of dogs over 10 years old in Serbia, as well as to determine some of the histochemical characteristics of these pigment and inclusions. O n L i n e F i r s t

MATERIALS AND METHODS
In this study, 36 dogs were necropsied. The dogs were randomly selected and divided into two groups (Table 1). The group of young dogs (group A) consisted of six dogs of different breeds and sexes (three males, three females) under five years old. The old dogs (group B) consisted of 13 males and 17 females over ten years old. During necropsy, the brain was removed, macroscopically examined and fixed in 10% neutral buffered formalin for six days. Coronal sections of the frontal cerebral cortex, parietal cerebral cortex, hippocampus, cerebellum and medulla oblongata were routinely processed, embedded in paraffin, cut into 4-6µm thick sections and stained using the haematoxylin and eosin (HE), long Ziehl Neelsen (Bancroft and Cook, 1984) and Periodic acid Schiff (PAS) methods. The stained tissues were analysed by light microscopy (Olympus, BX51) and sections with characteristic findings were captured by a digital camera (Olympus ColorView III). Legend: + (Lipofuscin in all examined segments of the brain, stained by one or both methods), -(lipofuscin not detected in any examined segments of brain, stained by both methods), P (PGBs in all examined segments of the the brain, extracellular accumulation), P * (PGBs in all examined segments of the brain, extracellular and intracellular accumulation), N (PGBs not detected in any of examined segments of the the brain), Fc (Frontal cortex), Pc (Parietal cortex), Hc (Hippocampus), C (Cerebellum), Mo (Medulla oblongata)

RESULTS
Lipofuscin occurred in several large neurons of the medulla oblongata of only one group A dog (Figure 1a and 2a), accumulated in a small amount in the cytoplasm and conventionally arranged around the nucleus in the form of fine granules, which stained red using PAS method. In group B dogs, tissue stained with HE showed the granular O n L i n e F i r s t structure of light brown to brown lipofuscin that was easily distinguished from the eosinophilic cytoplasm ( Figure 1b). The presence of lipofuscin was considerably more common in group B than in group A dog brains ( Figure 2). In all examined brain tissue segments, various percentages of the studied dogs had accumulated lipofuscin in neurons ( Figure 3) and lipofuscin granules were located perinuclearly as well as in the dendritic tree and the axon. The presence of lipofuscin pigment was demonstrated in large neurons of olivary nuclei of the medulla oblongata in 28 dogs (93.33%) using both methods of specific lipofuscin staining (Figure 2b and 4a). In addition to the medulla oblongata, in all other examined segments, the presence of lipofuscin granules in the neurons of group B dogs was confirmed using the long Ziehl Neelsen method (56% -74%) and the PAS staining method (60% -76%) (Figure 3).
PGBs were present as PAS-positive spheroids and were mainly located in the neuropil in all examined brain tissue segments ( Figure 5). Morphologically, they appeared, either as oval-shaped structures with central parts of a dark-purple to violet colour with light pink to pink halo, or as a homogeneous pink spheroid ( Figure 5). PGBs were demonstrated in almost all group B dogs, but they were not present in group A dogs ( Figure 6). PGBs in all examined brain tissue segments of group B dogs occurred   O n L i n e F i r s t intracellularly only in individual cases, in the hippocampus, parietal lobe, and in a cerebellum of dogs more than fifteen years old. Moreover, PGBs in the cerebellum, and frontal and parietal cortices were most often localised in molecular layers, while in the medulla oblongata, they were diffusely distributed. The number of PGBs depended on the age of the group B dogs. The presence of a large number of PGBs in the neuropil and intracellularly was observed in older dogs (more than 15 years old).

DISCUSSION
Some authors believe that dogs can be divided into three groups based on several criteria clinical evidence and morphological changes in the brain according to age. The three dog groups were: young dogs up to 5 years old, middle-aged dogs (5 to 10 years old) and old dogs (more than 10 years old). Dogs more than 10 years old show symptoms of cognitive dysfunction caused by morphological changes in the brain (Rofina et al., 2003, Youssef et al., 2016. In this study, the presence of lipofuscin in neurons from specific parts of the brain in dogs of different ages was examined. The results obtained with both specific staining methods indicated that lipofuscin was accumulated mostly in large neurons of the olivary nuclei of medulla oblongata (in 93% of dogs), while less lipofuscin occurred in other examined segments. Using both methods of specific staining on the same brain tissue segments, a small difference in the percentages of lipofuscin-positive individuals was observed. The percentage of positive cases stained by the long Ziehl Neelsen method ranged from 56% to 70%, while the PAS-positive cases ranged from 60% to 76% of all examined segments except for medulla oblongata. This indicates the PAS staining method is more sensitive for lipofuscin detection than the long Ziehl Neelsen staining method. O n L i n e F i r s t PAS staining of lipofuscin is thought to be due to insoluble lipid components, since PAS-positive substances are glycolipids or glycoproteins. The intensity of lipofuscin staining with carbol fuchsin increases with age, but the cause of this phenomenon is not yet clarified. It is known that acid resistance in Mycobacterium spp. is due to the high molecular weight of mycolic acid, which is composed of a large number of hydroxylated fatty acids. Therefore, the positivity of lipofuscin to Ziehl Neelsen staining could be a consequence of the presence of polymerised fatty acids (Benavides et al., 2002). We believe the long Ziehl Neelsen staining method was less sensitive to lipofuscin than the PAS method in our study precisely because of the polymerised fatty acid component of lipofuscin.
The amount of accumulated lipofuscin in the brain of group B dogs varied, but the most intensive accumulation was found in the oldest dogs, which is consistent with the literature data (Borras et al., 1999). We observed various amounts of accumulated lipofuscin in different brain segments, while the most significant accumulation was in the medulla oblongata, which is in accordance with other author's data (Double et al., 2008).
According to the literature, lipofuscin has been characterised as PAS-positive, acid resistant (stained by long Ziehl-Neelsen technique), argyrophilic and it can be stained with lipid stains such as Sudan Black and Oil-red-O. The tinctorial properties of lipofuscin increase in older dogs. Lipofuscin contains autofluorescent granules (golden yellow). In unstained tissue samples, lipofuscin can be visualised using fluorescence or laser scanning microscopy with the help of excitation light of different wavelengths and using different filters (Brunk and Terman, 2002b).
Similar to the literature data, the results of this study show the presence of PGBs in the brain of old dogs is a common finding, and the PGBs appear extracellularly and/ or intracellularly.
Some authors have noted PGBs are present in astrocytes as corpora amylacea (CAs), and in neurons as Lafora bodies. In the human population, CAs and Lafora bodies are normally present in the brain during the ageing process, while their number significantly increases in patients with Alzheimer's disease and myoclonic epilepsy (Borras et al., 1999, Lohi et al., 2005. The two types of PGBs observed in the human brain are well defined immunohistochemically. Unlike in humans, PGBs in animals are not fully defined, i.e., various results were obtained in numerous studies (Borras et al., 1999, Gredal et al., 2003. Immunohistochemical analysis of multiple markers showed different immuno-positive reactions in various brain segments. Therefore, some authors consider that PGBs in a cerebellum originate from neurons, and in other parts of the brain are derived from neurons and glial cells (Gredal et al., 2003).
In this study, the presence of PGBs in brain tissue was proven by PAS staining. PGBs occurred in the vast majority (93%) of old dogs, but were not observed in young dogs. Among the old dogs, PGBs were found in all examined brain tissue segments. Moreover, PGBs were observed extracellularly and intracellularly in the brains of the O n L i n e F i r s t oldest dogs, but in other dogs were only extracellular. Since in the examined brain segments of dogs, PGBs were primarily located extracellularly, they could not be corpora amylacea or Lafora bodies. Some authors have found that PAS-positive PGBs produced during the ageing process were ubiquitin-and neurofilament-negative. Taking into consideration the obtained results of double-staining by the PAS and immunohistochemical methods on glial fibrillary acid protein, as well as the absence of an immuno-positive reaction on ubiquitin in their research, other authors labelled these PGBs as Lafora bodies. The results obtained by double-staining indicated the PGBs were not related to the astrocytic processes (Borras et al., 1999).

CONCLUSION
The results obtained show, for the first time, the presence and characteristics of lipofuscin and PGBs in the brain of dogs more than 10 years old in Serbia. Lipofuscin occurred more frequently in the medulla oblongata than in other brain tissue segments.
Most PGBs in all examined brain tissue segments of dogs more than 10 years old were extracellular. The results show the changes we found in the old dog brain were similar to those reported in the literature. Also, the results obtained indicate some similarity between all the changes we found in the old dog brain and changes described as normal in the process of human ageing, and in patients with neurodegenerative disorders like Alzheimer's disease. This is one of numerous indications that old dogs could make a very good animal model for studying the normal ageing process or neurodegenerative diseases. Similarly, future research could give an answer as to what extent the relative age (age of the individual in regard to the lifespan of the breed) affects the development of changes in the brain and the effect of these changes on the individual's cognitive abilities.
Ključne reči: pas, proces starenja, mozak, lipofuscin, poliglukozanska tela O n L i n e F i r s t