This study demonstrated the characteristic molecular properties of PrP in several human endocrine and exocrine tissues, such as the pituitary, adrenal, and submandibular glands and the pancreas, which have different characteristics than the brain. Additionally, we detail the histological localization of PrP in these tissues.
Western blotting showed PrP smear bands ranging from 30 to 40 kDa in the pituitary gland, similar to previous reports5. These molecular weights are higher than those for the brain, and our previous study showed that this was due to excessive glycosylation. PrP signals were observed in the adrenal, pancreas, and submandibular glands at ~ 30–40 kDa and ~ 16–20 kDa. The ~ 30–40 kDa PrP smear signal was similar to that of the pituitary gland. However, a ~ 16–20 kDa PrP smear signal was not observed in the pituitary gland or the brain. The ~ 16–20 kDa PrP smear signal was recognized by the 3F4 antibody and the C-terminal antibody (EP1802Y), suggesting that it contains a C-terminal from the PrP core structure. Hachiya et al. showed that PrPc C-terminal fragments of approximately 20 kDa are present mainly on the cell surface19. However, the role of these PrPc C-terminal fragments in endocrine and exocrine tissues remains unclear.
Pit-1 activates transcription of the GH, prolactin, and beta-thyroid stimulating hormone genes20, and we found that most cytoplasmic PrPc positive adenohypophyseal cells were immunopositive for nuclear Pit-1. Conversely, cytoplasmic PrPc positive adenohypophyseal cells were essentially negative for SF-1 and TBX19. These findings suggest that the expression of PrPc in adenohypophyseal cells tents to occur in differentiated cells downstream from somatotroph stem cells rather than corticotroph cells and gonadotroph cells. A previous report showed that cytoplasmic PrPc positive adenohypophyseal cells are frequently positive for GH or prolactin5, and our results agree. However, there are no reports that suggest a direct association between PrP and Pit-1, and it is unclear when PrPc is highly expressed during cell differentiation by Pit-1.
In the adrenal gland, PrPc immunoreactivity was observed in the adrenal medulla. The adrenal medulla is derived from the neural crest, and PrPc is as abundant as in the central nervous system. For example, PrPSc was detected in the adrenal medulla in cattle affected with bovine spongiform encephalopathy and scrapie-infected sheep21, 22. In humans, we previously reported PrPSc aggregation in the adrenal medulla of sCJD patients and enhanced seeding activity on real-time quaking-induced conversion23. Our results suggested that PrPc in the adrenal medulla may be the source of PrPSc.
In the pancreas, islet cells have cytoplasmic PrP. In rats, PrPc is abundant in islet cells and is involved in regulating blood glucose homeostasis24. In addition, PrPc primarily expresses in pancreatic β cells in humans and may contribute to insulin resistance25, 26. CgA is a neuroendocrine protein localized to nerve cells and secretory vesicles of endocrine cells27. Reports have shown that PrPc and CgA colocalized in bovine islet cells.28 In our study, CgA and PrPc colocalization was noted in adenohypophyseal cells, pancreatic islet cells, and the adrenal medulla, suggesting that PrPc-expression in neuroendocrine cells could be expressed throughout the body of human. However, very high levels of CgA-expressing adenohypophyseal cells did not show cytoplasmic PrP. Therefore, even if the genes for CgA and PrPc are adjacent to each other, they do not necessarily colocalize in all tissues.
PrP immunopositivity was detected in the ductal epithelium of the submandibular gland and the pancreas. There have been no reports of PrP expression in the ductal epithelium of animals or humans. CWD is a fatal neurodegenerative prion disease that affects species of the Cervidae family and is now a widespread global problem in the United States, Canada, and Europe12, 29, 30. Saliva and feces containing PrPSc are likely sources of infection29, 30. Our results suggest that in CWD animals, PrPSc may be secreted from the ductal epithelium in saliva and from the pancreas through the gastrointestinal tract to feces.
This study detected PrPc in human endocrine and exocrine tissues using western blotting and clarified the histological details. PrPc is an element of the crinophagy mechanism (an autophagy mechanism that processes excess hormones and neurotransmitters) in secretory cells. Furthermore, PrPc is involved in the secretory pathway of secretory granules.8 The immunohistochemistry and western blotting detection sensitivities for PrPc varied between cases. This may be due to different endocrine and exocrine function states per case during the period before death. The immunohistochemistry detection sensitivities also differed depending on the PrP antibody, which may be related to antigen exposure owing to the three-dimensional structure of PrPc. This study includes several limitations. Frozen samples of endocrine and exocrine tissue could not be collected adequately at autopsy, therefore, analysis of western blots was performed from FFPE samples. It is undeniable that modification of the sample during the FFPE manufacturing process can affect the results of western blot. However, western blot results from FFPE of the brain and pituitary gland were similar to those of frozen samples. In addition, since it is a biochemical result of the tissue adjacent to the specimen used for histological observation, we think that it is also a very suitable method for examining the tissue localization of PrPc. Therefore, we believe that the characteristic signals found in western blots of the adrenal glands, pancreas and submandibular glands are reliable.