Portal de Periódicos da CAPES

Subcellular Localization of Disease-Associated Prion Protein in the Human Brain

2005; Elsevier BV; Volume: 166; Issue: 1 Linguagem: Inglês

10.1016/s0002-9440(10)62252-3

1525-2191

Gábor G. Kovács, Matthias Preusser, Michaela Strohschneider, Herbert Budka,

Trace Elements in Health

Disease-associated prion protein (PrPTSE) deposits in distinct immunostaining patterns in the brain in Creutzfeldt-Jakob disease, including synaptic, extracellular, and cell-associated localizations. After having developed an appropriate pretreatment protocol to enhance immunostaining for PrPTSE without damaging epitopes of other antigens, we systematically evaluated co-localization patterns of distinct PrPTSE immunodeposits by confocal laser microscopy, including optical serial sectioning. As shown by quantification, the most prominent co-localization of PrPTSE is with synaptophysin, but PrPTSE may also co-deposit with connexin-32, a gap junction-related protein. Furthermore, neuronal cell bodies, dendrites, axons, astrocytes, and microglia harbor granular PrPTSE deposits. Highly aggregated deposits are focally ubiquitinated. We conclude that PrPTSE is not exclusively associated with chemical but also with electric synapses, axonal transport may be a relevant route of PrPTSE spread in the brain, and activated microglia and astrocytes may play a role in PrPTSE processing, degradation, or removal. Disease-associated prion protein (PrPTSE) deposits in distinct immunostaining patterns in the brain in Creutzfeldt-Jakob disease, including synaptic, extracellular, and cell-associated localizations. After having developed an appropriate pretreatment protocol to enhance immunostaining for PrPTSE without damaging epitopes of other antigens, we systematically evaluated co-localization patterns of distinct PrPTSE immunodeposits by confocal laser microscopy, including optical serial sectioning. As shown by quantification, the most prominent co-localization of PrPTSE is with synaptophysin, but PrPTSE may also co-deposit with connexin-32, a gap junction-related protein. Furthermore, neuronal cell bodies, dendrites, axons, astrocytes, and microglia harbor granular PrPTSE deposits. Highly aggregated deposits are focally ubiquitinated. We conclude that PrPTSE is not exclusively associated with chemical but also with electric synapses, axonal transport may be a relevant route of PrPTSE spread in the brain, and activated microglia and astrocytes may play a role in PrPTSE processing, degradation, or removal. Post-translational conversion of the α-helix-dominant cellular prion protein (PrPC) into the β-sheet-dominant PrPSc is a common characteristic of animal and human prion diseases including Creutzfeldt-Jakob disease (CJD).1Prusiner SB Prions.Proc Natl Acad Sci USA. 1998; 95: 13363-13383Crossref PubMed Scopus (5245) Google Scholar Both protease-resistant and protease-sensitive forms of PrPSc are known.2Safar J Wille H Itri V Groth D Serban H Torchia M Cohen FE Prusiner SB Eight prion strains have PrP(Sc) molecules with different conformations.Nat Med. 1998; 4: 1157-1165Crossref PubMed Scopus (1100) Google Scholar Due to the lack of unequivocally conformation-specific antibodies, immunohistochemical differentiation of PrPC from PrPSc has remained a matter of debate and creates a barrier for the evaluation of disease process. However, detailed immunohistochemical descriptions define patterns which are specific for disease,3Kovács GG Head MW Hegyi I Bunn TJ Flicker H Hainfellner JA McCardle L Laszlo L Jarius C Ironside JW Budka H Immunohistochemistry for the prion protein: comparison of different monoclonal antibodies in human prion disease subtypes.Brain Pathol. 2002; 12: 1-11Crossref PubMed Scopus (101) Google Scholar thus the term disease-associated PrP (labeled PrPTSE in the present study) is suitable to label conformationally uncharacterized deposits of PrP observed only in prion-diseased brains by immunohistochemistry. In the diseased human brain, main deposition patterns of PrPTSE are so-called diffuse/synaptic, patchy/perivacuolar, perineuronal, and plaque-like patterns.3Kovács GG Head MW Hegyi I Bunn TJ Flicker H Hainfellner JA McCardle L Laszlo L Jarius C Ironside JW Budka H Immunohistochemistry for the prion protein: comparison of different monoclonal antibodies in human prion disease subtypes.Brain Pathol. 2002; 12: 1-11Crossref PubMed Scopus (101) Google Scholar The first is characterized by a diffuse fine-granular distribution of anti-PrPTSE immunoreactivity resembling the distribution of the presynaptic protein synaptophysin (hence the term "synaptic pattern"). Presynaptic domains appear to be the privileged site of PrPC, as has been confirmed by double labeling with synaptophysin.4Fournier JG Escaig-Haye F Grigoriev V Ultrastructural localization of prion proteins: physiological and pathological implications.Microsc Res Tech. 2000; 50: 76-88Crossref PubMed Scopus (82) Google Scholar Also a post-synaptic localization of PrPC is sometimes observed in synapses of the central nervous system (CNS) and muscle-nerve synapses.4Fournier JG Escaig-Haye F Grigoriev V Ultrastructural localization of prion proteins: physiological and pathological implications.Microsc Res Tech. 2000; 50: 76-88Crossref PubMed Scopus (82) Google Scholar Others suggest that PrPC is almost excluded from synaptic vesicles and also describe its presence in the cytosol of certain neuronal subpopulations.5Mironov AJ Latawiec D Wille H Bouzamondo-Bernstein E Legname G Williamson RA Burton D DeArmond SJ Prusiner SB Peters PJ Cytosolic prion protein in neurons.J Neurosci. 2003; 23: 7183-7193Crossref PubMed Google Scholar Cell culture studies indicated a destiny of PrPC for the plasma membrane and caveolae-like membranous domains.6Vey M Pilkuhn S Wille H Nixon R DeArmond SJ Smart EJ Anderson RG Taraboulos A Prusiner SB Subcellular colocalization of the cellular and scrapie prion proteins in caveolae-like membranous domains.Proc Natl Acad Sci USA. 1996; 93: 14945-14949Crossref PubMed Scopus (492) Google Scholar Immunogold electron microscopy has demonstrated presence of PrP epitopes in a presynaptic localization in human variant CJD and bovine spongiform encephalopathy-infected monkey brain.4Fournier JG Escaig-Haye F Grigoriev V Ultrastructural localization of prion proteins: physiological and pathological implications.Microsc Res Tech. 2000; 50: 76-88Crossref PubMed Scopus (82) Google Scholar, 7Grigoriev V Escaig-Haye F Streichenberger N Kopp N Langeveld J Brown P Fournier JG Submicroscopic immunodetection of PrP in the brain of a patient with a new-variant of Creutzfeldt-Jakob disease.Neurosci Lett. 1999; 264: 57-60Crossref PubMed Scopus (43) Google Scholar Additionally, subcellular fractionation has shown that the highest concentration of PrPTSE in CJD brain is found in the synaptosomal fraction, however, without distinguishing between pre- and post-synaptic localization.8Kitamoto T Shin RW Doh-ura K Tomokane N Miyazono M Muramoto T Tateishi J Abnormal isoform of prion proteins accumulates in the synaptic structures of the central nervous system in patients with Creutzfeldt-Jakob disease.Am J Pathol. 1992; 140: 1285-1294PubMed Google Scholar Experimental data suggest that PrPTSE may also be found in other parts of neurons than their synaptic terminals, furthermore, there is also evidence for accumulation of PrPTSE in astrocytes and microglia.4Fournier JG Escaig-Haye F Grigoriev V Ultrastructural localization of prion proteins: physiological and pathological implications.Microsc Res Tech. 2000; 50: 76-88Crossref PubMed Scopus (82) Google Scholar Due to methodological difficulties, confirmatory evaluation of PrPTSE deposition in the diseased human brain is lacking. After having developed an appropriate pretreatment protocol to visualize immunostaining for PrPTSE without damaging epitopes of other antigens, we systematically studied PrPTSE deposition by double immunofluorescent labeling and evaluated co-localization using laser confocal microscopy. Four autopsy cases of sporadic Creutzfeld-Jakob disease (sCJD), with full-length analysis of the prion protein gene (PRNP), were selected from the tissue bank of the Institute of Neurology, Medical University Vienna, Austria. We selected CJD cases with relatively preserved cortical structure along with high amounts of immunohistochemically detectable PrPTSE. Cases 1 (72-year-old female, 5-month disease duration) and 2 (65-year-old female, 1-month disease duration) were methionine/methionine (MM) homozygotes at the polymorphic codon 129 of PRNP; case 3 (74-year-old male, 6-week disease duration) was a VV homozygote; and case 4 (67-year-old female, 12-month disease duration) was a methionine/valine (MV) heterozygote. Two control cases without prion disease were included as well (control case 1: 78-year-old male, cause of death, myocardiac insufficiency; control case 2: 78-year-old female, cause of death, myocardiac infarction). In all cases, brain tissue samples containing cerebral cortex and white matter were routinely fixed in 4% buffered formalin. This was performed within 12 hours (CJD cases 2 and 3), 24 hours (CJD case 4 and control case 1), 48 hours (CJD case 1 and control case 2) after death. Tissues were stored in formalin for 5 days (CJD case 1), 5 weeks (CJD case 2), 2 months (CJD case 3), 6 days (CJD case 4), 11 days (control case 1), and 4 days (control case 2) before paraffin-embedding. Paraffin-embedded tissue blocks were stored in darkness and at room temperature for 36 months (CJD case 1), 1 month (CJD case 2), 18 months (CJD case 3), 1 month (CJD case 4), 2 months (control case 1), and 2 weeks (control case 2) before being processed for this study as specified below. Five-μm thick sections were obtained from each tissue block for conventional immunohistochemistry and double immunofluorescent labeling. Twenty-μm thick tissue sections were used for optical serial sectioning ("z-stacking"). After a pilot fashion testing of different pretreatment protocols suitable to detect both PrPTSE and selected antigens, we defined a protocol that included 20 minutes of autoclaving (100°C) in citrate buffer (pH 6.0) and incubation for 2 minutes in formic acid (96%). Commercially available primary antibodies with well-defined specificity used for this study are listed in Table 1. We used the monoclonal anti-PrP antibody 12F10 that was previously shown to be the most reliable in detecting disease-associated PrP.3Kovács GG Head MW Hegyi I Bunn TJ Flicker H Hainfellner JA McCardle L Laszlo L Jarius C Ironside JW Budka H Immunohistochemistry for the prion protein: comparison of different monoclonal antibodies in human prion disease subtypes.Brain Pathol. 2002; 12: 1-11Crossref PubMed Scopus (101) Google Scholar Sections were incubated overnight at room temperature with each antibody.Table 1Summary of Antibodies Used in This StudyAntibodySupplierDilutionClonalitySpecifity (labeled structures)12F10Cayman Chemicals, Ann Arbor, MI1:300Mono (Mouse) IgG2aPrPSMI-31Sternberger Mo. Ltd., Lutherville, MD1:2500Mono (Mouse) IgG1Phosphorylated epitope in extensively phosphorylated neurofilament H and, to a lesser extent, with neurofilament M. Reacts broadly with thick and thin axons and some dendrites such as basket cell dendrites, but not Purkinje cell dendrites. Nerve cell bodies are generally unreactive.SMI-32Sternberger Mo. Ltd., Lutherville, MD1:100Mono (Mouse) IgG1Non-phosphorylated epitopes in neurofilament H. Reacts with neuronal cell bodies and dendrites and only some thick axons.AP20Chemicon, Temecula, CA1:250tMono (Mouse) IgG1Microtubule associated protein-2 (dendrites + neuronal body and occasional axons)SY38Dako, Glostrup, Denmark1:10Mono (Mouse) IgG1Synaptophysin (presynaptic structures, synaptic vesicle)Anti-synapsin IAlexis Biochemicals, Grünberg, Germany1:25Poly (Rabbit)Synapsin I (synaptic vesicles)M12.13Zymed, San Francisco, CA1:25Mono (Mouse) IgGConnexin-32 (gap junction protein-electrical synapses)Anti-GFAPDako, Glostrup, Denmark1:2000Poly (Rabbit)Glial fibrillar acidic protein (reactive astroglia)CR3/43Dako, Glostrup, Denmark1:50Mono (Mouse) IgG1HLA-DR (reactive microglia)Anti-ubiquitinDako, Glostrup, Denmark1:100Poly (Rabbit)Ubiquitin (ubiquitinated protein aggregates) Open table in a new tab For double immunolabeling, the fluorescent-labeled secondary antibody for the anti-PrP antibody 12F10 was Alexa Fluor 488 goat anti-mouse IgG (1:200, Molecular Probes, USA), for anti-SMI-31, anti-SMI-32, anti-MAP-2, anti-connexin-32, anti-synaptophysin, and anti-HLA-DR Alexa Fluor 546 goat anti-mouse IgG (1:200, Molecular Probes), and for synapsin, GFAP, and ubiquitin goat anti-rabbit IgG (1:200, Molecular Probes). Primary antibodies were visualized with fluorophore-labeled, anti-mouse Fc, Fab fragments, using the Zenon Alexa Fluor 488 Mouse IgG2a Labeling Kit (Molecular Probes, Leiden, The Netherlands) for 12F10, and the Zenon Alexa Fluor 546 Mouse IgG1 Labeling Kit (Molecular Probes, Leiden, The Netherlands) for other antibodies. We used argon 488-nm and helium/neon 633-nm and 543-nm lasers to elicit immunofluorescent staining. Immunofluorescence labeling was evaluated using a Zeiss LSM 510 laser scanning confocal microscope. The optical pinhole was adjusted for each scanning experiment to obtain optical slices with a 0.2-μm thickness. Optical serial sections with distances of 0.5 μm between individual sections were obtained to confirm intracellular localization of PrPTSE deposits. To compare the proportion of co-localized pixels of PrPTSE/synaptophysin and PrPTSE/connexin-32 we extracted 10 randomly selected 1000 μm2 areas for quantification from the scanned images of double immunolabeling. Accepting bright yellow color as a marker of co-labeling of green and red particles, we used Adobe Photoshop 6.0 to select one bright yellow-colored pixelbox from each of the saved images. Pixelboxes containing bright yellow components were highlighted and counted manually on the screen. A similar procedure was performed for green color. We then calculated the percentage of pixelboxes containing a yellow component from the total number (yellow+green) of counted pixelboxes. By testing pretreatment protocols, we have defined a protocol suitable for our study. This enabled us to detect either diffuse/synaptic, patchy/perivacuolar, perineuronal, or plaque-type PrPTSE deposits sufficiently after this pretreatment (Figure 1, a to d). In comparison with our standard immunohistochemical pretreatment protocol,3Kovács GG Head MW Hegyi I Bunn TJ Flicker H Hainfellner JA McCardle L Laszlo L Jarius C Ironside JW Budka H Immunohistochemistry for the prion protein: comparison of different monoclonal antibodies in human prion disease subtypes.Brain Pathol. 2002; 12: 1-11Crossref PubMed Scopus (101) Google Scholar the signal was slightly less intense but labeled all structures similarly (Figure 1, m and n). In addition, all other antibodies used in our study convincingly detected their respective antigens with this protocol (Figure 1, e to l) in contrast to the loss of immunoreactivity after using the three-tiered pretreatment protocol for PrPTSE (Figure 1, o to t). In normal control brains we found no immunoreactivity for PrPTSE, and we did not observe immunolabeling of any cells in the CNS (Figure 4).Figure 4Confocal laser scanning microscopy of double immunofluorescent labeling with 12F10 (a–d) and anti-GFAP (a), SMI 31 (b) anti-synaptophysin (c), and anti-connexin-32 using the modified pretreatment protocol in a non-CJD brain (control case 1; blue signals indicate autofluorescence, eg, lipopigment). No green PrP deposits are detectable, while astrocytes (a), axons (b), presynaptic structures (c), and gap junction-related components (d) are immunolabeled (red).View Large Image Figure ViewerDownload Hi-res image Download (PPT) In all cases, fine granular, so-called diffuse/synaptic type, of PrPTSE deposits co-localized with synaptophysin (Figure 2a). A similar co-labeling pattern was seen with synapsin I (data not shown). Quantification analysis revealed co-localization of PrPTSE with synaptophysin in 72.91 + 5.9% of pixels. Coarse PrPTSE deposits like patchy/perivacuolar or plaque type did not show co-localization with synaptophysin. Co-labeling of fine granular PrPTSE and connexin-32 was shown in 10.78 + 3.8% of pixels (Figure 2b). Using the SMI-31 antibody to immunolabel axons, we noted co-localization with PrPTSE within the axon (Figure 2, c to e). This was also evident after creating split images where the incorporated PrPTSE granular deposits follow the outline of an axon (Figure 2c). Intra-axonal localization of PrPTSE granules was confirmed by optical serial sections showing that individual PrPTSE granules are surrounded by axoplasm on all sides (Figure 2, d and e). Infrequently, we observed clustering of granular PrPTSE immunoreactivity within the perikaryon in close association with the surface of neurons (Figure 2f) and dendrites (Figure 2g) of neurons. Additionally there is perineuronal and periaxonal deposition of PrPTSE (Figure 2f). None of the neuronal markers showed co-labeling with PrPTSE in aggregated patchy/perivacuolar deposits. Rarely, we observed PrPTSE deposits within micro- (Figure 3, a and b) and astroglial (Figure 3, c to e) cells. Similarly to the intra-axonal distribution, intracytoplasmic astro- and microglial localization of PrPTSE deposits was confirmed by optical serial sectioning, showing that PrPTSE deposits are surrounded by cytoplasm on all sides. Occasionally PrPTSE deposits were clustered in close vicinity of astroglial cells that also showed intracytoplasmic accumulation of PrPTSE (Figure 3e). Parts of extensive aggregated deposits of PrPTSE co-localized with ubiquitin frequently (Figure 2f) while fine granular deposits do not appear to be ubiquitinated. Immunohistochemistry for PrP is a powerful tool for diagnosing prion disease.9Budka H Histopathology and immunohistochemistry of human transmissible spongiform encephalopathies (TSEs).Arch Virol Suppl. 2000; 16: 135-142PubMed Google Scholar To avoid misinterpretation caused by the inability of available antibodies to distinguish between normal and pathological conformers of PrP, tissues must undergo special pretreatment before immunostaining to abolish PrPC and to enhance PrPTSE immunoreactivities. Commonly used pretreatment protocols use the combination of hydrated or hydrolytic autoclaving at 121° C of formic acid, guanidine thiocyanate, and proteinase K.3Kovács GG Head MW Hegyi I Bunn TJ Flicker H Hainfellner JA McCardle L Laszlo L Jarius C Ironside JW Budka H Immunohistochemistry for the prion protein: comparison of different monoclonal antibodies in human prion disease subtypes.Brain Pathol. 2002; 12: 1-11Crossref PubMed Scopus (101) Google Scholar However, these pretreatment protocols also damage other antigens present in the tissue, making their immunodetection impossible. This is one reason for the paucity of co-labeling studies defining the exact localization of PrPTSE in the CNS. Regional distribution of PrPTSE was shown to be different in sporadic, iatrogenic, and genetic prion disease cases.10Brown P Kenney K Little B Ironside JW Will R Cervenakova L Bjork RJ San Martin RA Safar J Roos R Haltia M Gibbs CJ Gajdusek DC Intracerebral distribution of infectious amyloid protein in spongiform encephalopathy.Ann Neurol. 1995; 38: 245-253Crossref PubMed Scopus (42) Google Scholar Later molecular classification based on Western blot characteristics of the protease-resistant PrP and the PRNP polymorphism at codon 129 has revealed that the PrP immunostaining patterns are distinguishable in anatomical regions and molecular phenotypes of sporadic CJD cases.11Parchi P Giese A Capellari S Brown P Schulz-Schaeffer W Windl O Zerr I Budka H Kopp N Piccardo P Poser S Rojiani A Streichemberger N Julien J Vital C Ghetti B Gambetti P Kretzschmar H Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects.Ann Neurol. 1999; 46: 224-233Crossref PubMed Scopus (1247) Google Scholar, 12Hill AF Joiner S Wadsworth JD Sidle KC Bell JE Budka H Ironside JW Collinge J Molecular classification of sporadic Creutzfeldt-Jakob disease.Brain. 2003; 126: 1333-1346Crossref PubMed Scopus (282) Google Scholar In our present study we collected samples harboring all major morphological subtypes of PrPTSE immunodeposits related to sporadic CJD brains. Co-localization of PrPTSE deposits with synaptophysin and synapsin I confirms previous ultrastructural studies that PrPTSE is present in presynaptic components. Quantification analysis, however, suggests the existence of other sites of PrPTSE deposition. Surprisingly, we observed co-labeling of fine granular PrPTSE and connexin-32. Connexins are molecular constituents of gap junctions (electric synapses) that are distributed in astrocytes, oligodendrocytes, and neurons.13Bennett MV Barrio LC Bargiello TA Spray DC Hertzberg E Saez JC Gap junctions: new tools, new answers, new questions.Neuron. 1991; 6: 305-320Abstract Full Text PDF PubMed Scopus (890) Google Scholar, 14Rouach N Avignone E Meme W Koulakoff A Venance L Blomstrand F Giaume C Gap junctions and connexin expression in the normal and pathological central nervous system.Biol Cell. 2002; 94: 457-475Crossref PubMed Scopus (292) Google Scholar Connexin-32 is predominantly related to neurons, less to oligo-, and not to astrocytes.14Rouach N Avignone E Meme W Koulakoff A Venance L Blomstrand F Giaume C Gap junctions and connexin expression in the normal and pathological central nervous system.Biol Cell. 2002; 94: 457-475Crossref PubMed Scopus (292) Google Scholar Interestingly, mainly GABAergic interneurons are interconnected through electric synapses, a subset of which is particularly vulnerable in human and experimental prion diseases.15Guentchev M Hainfellner JA Trabattoni GR Budka H Distribution of parvalbumin-immunoreactive neurons in brain correlates with hippocampal and temporal cortical pathology in Creutzfeldt-Jakob disease.J Neuropathol Exp Neurol. 1997; 56: 1119-1124Crossref PubMed Scopus (75) Google Scholar, 16Guentchev M Groschup MH Kordek R Liberski PP Budka H Severe, early and selective loss of a subpopulation of GABAergic inhibitory neurons in experimental transmissible spongiform encephalopathies.Brain Pathol. 1998; 8: 615-623Crossref PubMed Scopus (75) Google Scholar Accumulation of PrPTSE in gap junctions may be a substrate for neuronal damage as the role of these structures involves intra- and extracellular homeostasis, trafficking and supply of energy-producing substrates to neurons, or neuroprotection.14Rouach N Avignone E Meme W Koulakoff A Venance L Blomstrand F Giaume C Gap junctions and connexin expression in the normal and pathological central nervous system.Biol Cell. 2002; 94: 457-475Crossref PubMed Scopus (292) Google Scholar We observed unequivocal localization of PrPTSE granules within axons. Previous ultrastructural studies have suggested association of PrPTSE with the plasmalemma of neurites,17Jeffrey M Goodsir CM Bruce ME McBride PA Scott JR Halliday WG Infection-specific prion protein (PrP) accumulates on neuronal plasmalemma in scrapie-infected mice.Neurosci Lett. 1992; 147: 106-109Crossref PubMed Scopus (68) Google Scholar, 18Jeffrey M Goodsir CM Bruce ME McBride PA Fowler N Scott JR Murine scrapie-infected neurons in vivo release excess prion protein into the extracellular space.Neurosci Lett. 1994; 174: 39-42Crossref PubMed Scopus (43) Google Scholar however, PrPTSE has never been detected inside neurites. Presence of PrPTSE in the cytosol of axons is not unexpected, since a recent study demonstrated the existence of cytosolic PrPC probably following a retro-translocation from the endoplasmic reticulum.5Mironov AJ Latawiec D Wille H Bouzamondo-Bernstein E Legname G Williamson RA Burton D DeArmond SJ Prusiner SB Peters PJ Cytosolic prion protein in neurons.J Neurosci. 2003; 23: 7183-7193Crossref PubMed Google Scholar As PrPTSE has been shown to traffic also in endosome-like organelles,19Arnold JE Tipler C László L Hope J Landon M Mayer RJ The abnormal isoform of the prion protein accumulates in late-endosome-like organelles in scrapie-infected mouse brain.J Pathol. 1995; 176: 403-411Crossref PubMed Scopus (150) Google Scholar it is also conceivable that our demonstration of intra-axonal PrPTSE has such subcellular basis. However, our observation merits (immuno)-ultrastructural confirmation to clarify whether PrPTSE within the axons is free in the cytosol or related to vesicles (eg, endosomal). Recent observations in experimental models suggest that transport of PrPTSE along peripheral nerves to the CNS is following a domino-like manner along PrPC-expressing nerve membranes rather than being achieved via conventional axonal transport.20Kunzi V Glatzel M Nakano MY Greber UF Van Leuven F Aguzzi A Unhampered prion neuroinvasion despite impaired fast axonal transport in transgenic mice overexpressing four-repeat tau.J Neurosci. 2002; 22: 7471-7477Crossref PubMed Google Scholar In human prion disease PrPTSE deposits in the CNS may be observed in both fine and coarse perineuronal patterns supporting a domino-like manner of propagation. However, in human disease the perineuronal pattern is not predominant. It is seen mainly in VV or MV type 2 subtypes of sporadic CJD cases and variant CJD, suggesting that this might be a strain-dependent phenomenon.3Kovács GG Head MW Hegyi I Bunn TJ Flicker H Hainfellner JA McCardle L Laszlo L Jarius C Ironside JW Budka H Immunohistochemistry for the prion protein: comparison of different monoclonal antibodies in human prion disease subtypes.Brain Pathol. 2002; 12: 1-11Crossref PubMed Scopus (101) Google Scholar, 21Ironside JW Pathology of variant Creutzfeldt-Jakob disease.Arch Virol Suppl. 2000; 16: 143-151PubMed Google Scholar, 22Kovács GG Head MW Bunn T Laszlo L Will RG Ironside JW Clinicopathological phenotype of codon 129 valine homozygote sporadic Creutzfeldt-Jakob disease.Neuropathol Appl Neurobiol. 2000; 26: 463-472Crossref PubMed Scopus (38) Google Scholar In contrast, fine granular, so-called diffuse/synaptic, deposits of PrPTSE represent the major morphological appearance. We thus suggest that PrPTSE in the CNS follows at least two pathways, one along and one within neurites. Although PrPC is subject to fast anterograde transport,23Rodolfo K Hassig R Moya KL Frobert Y Grassi J Di Giamberardino L A novel cellular prion protein isoform present in rapid anterograde axonal transport.Neuroreport. 1999; 10: 3639-3644Crossref PubMed Scopus (36) Google Scholar other possibilities need to be elucidated. One important route is probably ad-axonal, however, in circumstances with overwhelming production of PrPTSE, which is more likely to happen in the CNS and not in the periphery, conventional axonal transport systems may participate in distribution of PrPTSE. Clusters of granular PrPTSE immunoreactivity within the perikaryon in close association with the surface of neurons and dendrites of neurons must be distinguished from an occasional diffuse cytoplasmic immunolabeling of neurons which has also been described in non-diseased brains and is considered to represent cellular PrP.24Kovács GG Voigtlander T Hainfellner JA Budka H Distribution of intraneuronal immunoreactivity for the prion protein in human prion diseases.Acta Neuropathol. 2002; 104: 320-326PubMed Google Scholar, 25McKinley MP Taraboulos A Kenaga L Serban D Stieber A DeArmond SJ Prusiner SB Gonatas N Ultrastructural localization of scrapie prion proteins in cytoplasmic vesicles of infected cultured cells.Lab Invest. 1991; 65: 622-630PubMed Google Scholar The anti-PrP antibody (12F10) applied in our present study does not show this type of diffuse neuronal labeling.3Kovács GG Head MW Hegyi I Bunn TJ Flicker H Hainfellner JA McCardle L Laszlo L Jarius C Ironside JW Budka H Immunohistochemistry for the prion protein: comparison of different monoclonal antibodies in human prion disease subtypes.Brain Pathol. 2002; 12: 1-11Crossref PubMed Scopus (101) Google Scholar The granular deposits in the cell body and dendrites suggest a post-synaptic localization of PrPTSE. This is compatible with previous descriptions of a post-synaptic localization of cellular PrP,4Fournier JG Escaig-Haye F Grigoriev V Ultrastructural localization of prion proteins: physiological and pathological implications.Microsc Res Tech. 2000; 50: 76-88Crossref PubMed Scopus (82) Google Scholar which is considered to be a prerequisite for formation of PrPTSE. We confirm previous ultrastructural studies that have already noted PrPTSE in neuronal perikarya that was proposed to be mainly related to the lysosomal-endosomal system there.4Fournier JG Escaig-Haye F Grigoriev V Ultrastructural localization of prion proteins: physiological and pathological implications.Microsc Res Tech. 2000; 50: 76-88Crossref PubMed Scopus (82) Google Scholar The PrPC-PrPSc conversion was suggested to use the endocytic pathway via the caveola-like domain membrane or coated pits to the endosomal-lysosomal system. The latter were suggested to be key organelles in the processing route toward production of PrPSc not only in cell culture, but also in animal TSE and human variant CJD brains.7Grigoriev V Escaig-Haye F Streichenberger N Kopp N Langeveld J Brown P Fournier JG Submicroscopic immunodetection of PrP in the brain of a patient with a new-variant of Creutzfeldt-Jakob disease.Neurosci Lett. 1999; 264: 57-60Crossref PubMed Scopus (43) Google Scholar, 26László L Lowe J Self T Kenward N Landon M McBride T Farquhar C McConnell I Brown J Hope J Mayer RJ Lysosomes as key organelles in the pathogenesis of prion encephalopathies.J Pathol. 1992; 166: 333-341Crossref PubMed Scopus (157) Google Scholar Having found no neuronal markers within aggregated patchy/perivacuolar or plaque-like deposits we excluded the possibility that neuronal debris is a component of these amorphous materials. Our observations that micro- and astroglial cells harbor intracytoplasmic PrPTSE granules and that there is periastrocytic accumulation of PrPTSE deposits suggest that both microglial and astroglial cells may have a role in the processing, degradation, or removal of PrPTSE. Whether microglia are capable of phagocytosing aggregated PrPTSE material has rarely been addressed in research, although prominent microgliosis is a characteristic finding in CJD brains. Only a recent study demonstrated that spleen macrophages actively participate in the clearance of intraperitoneal scrapie inoculum.27Beringue V Demoy M Lasmezas CI Gouritin B Weingarten C Des JP Andreux JP Couvreur P Domont D Role of spleen macrophages in the clearance of scrapie agent early in pathogenesis.J Pathol. 2000; 190: 495-502Crossref PubMed Scopus (128) Google Scholar Activation of micro- and astroglial cells at sites of aggregated PrPTSE deposits and accumulation of PrPTSE in lysosomes have been documented previously.28Jeffrey M Goodsir CM Bruce ME McBride PA Farquhar C Morphogenesis of amyloid plaques in 87V murine scrapie.Neuropathol Appl Neurobiol. 1994; 20: 535-542Crossref PubMed Scopus (70) Google Scholar In an in vitro model using prion fragment 106–126 it has been shown that microglia is needed for neurotoxicity, while the same fragment stimulates proliferation of astrocytes.29Perry VH Cunningham C Boche D Atypical inflammation in the central nervous system in prion disease.Curr Opin Neurol. 2002; 15: 349-354Crossref PubMed Scopus (150) Google Scholar, 30Forloni G Del Bo R Angeretti N Chiesa R Smiroldo S Doni R Ghibaudi E Salmona M Porro M Verga L Giaccone G Bugiani O Tagliavini F A neurotoxic prion protein fragment induces rat astroglial proliferation and hypertrophy.Eur J Neurosci. 1994; 6: 1415-1422Crossref PubMed Scopus (107) Google Scholar Microglia activation occurs early in disease but there is neither direct in vivo evidence that PrPSc is proinflammatory nor was it demonstrated to what microglial cells actually respond.29Perry VH Cunningham C Boche D Atypical inflammation in the central nervous system in prion disease.Curr Opin Neurol. 2002; 15: 349-354Crossref PubMed Scopus (150) Google Scholar Ubiquitination of PrPTSE was already suggested by demonstrating ubiquitin immunoreactivity in areas where PrPTSE deposits are observed (eg, as perivacuolar immunopositivity).31Ironside JW McCardle L Hayward PA Bell JE Ubiquitin immunocytochemistry in human spongiform encephalopathies.Neuropathol Appl Neurobiol. 1993; 19: 134-140Crossref PubMed Scopus (61) Google Scholar Here we confirm that highly aggregated PrPTSE deposits indeed co-localize with ubiqutin frequently. That disease-associated PrP may be ubiquitinated is supported by a recent study, using immunoprecipitation and sandwich ELISA, showing ubiqituination of PrP after developing protease resistance in the brains of scrapie-infected mice.32Kang S Brown DR Whiteman M Li R Pan T Perry G Wisniewski T Sy M Wong B Prion protein is ubiquitinated after developing protease resistance in the brains of scrapie-infected mice.J Pathol. 2003; 203: 603-608Crossref Scopus (46) Google Scholar In sum, our study suggests alteration of both chemical, indicated by the presynaptic protein synaptophysin, and electric synapses, indicated by the gap junction-related protein connexin-32, by accumulated PrPTSE. This might be a central substrate of prion disease pathogenesis and clinical deterioration either by abolishing homeostatic communication between cells or by releasing inhibition. By demonstrating PrPTSE in the neuronal cell body and dendrites, a role of post-synaptic, in addition to previously assumed presynaptic, structures in CJD pathogenesis is proposed. The observation of both intra- and ad-axonal localization of PrPTSE suggests that both intra-axonal transport as well as domino-like propagation along the axonal surface may occur in the CNS. Moreover, our results suggest that PrPTSE may be processed or degraded in micro- or astroglial cells. We thank Gerda Ricken for excellent technical assistance.