Anatomic survey of seeding in Alzheimer’s disease brains reveals unexpected patterns

Barbara E. Stopschinski; Kelly Del Tredici; Sandi-Jo Estill-Terpack; Estifanos Ghebremdehin; Fang F. Yu; Heiko Braak; Marc I. Diamond

doi:10.1186/s40478-021-01255-x

Anatomic survey of seeding in Alzheimer’s disease brains reveals unexpected patterns

Stopschinski, Barbara E.; Del Tredici, Kelly; Estill-Terpack, Sandi-Jo; Ghebremdehin, Estifanos; Yu, Fang F.; Braak, Heiko; Diamond, Marc I. 2021-10-11 00:00:00 Tauopathies are heterogeneous neurodegenerative diseases defined by progressive brain accumulation of tau aggre ‑ gates. The most common tauopathy, sporadic Alzheimer’s disease (AD), involves progressive tau deposition that can be divided into specific stages of neurofibrillary tangle pathology. This classification is consistent with experimental data which suggests that network‑based propagation is mediated by cell–cell transfer of tau “seeds”, or assemblies, that serve as templates for their own replication. Until now, seeding assays of AD brain have largely been limited to areas previously defined by NFT pathology. We now expand this work to additional regions. We selected 20 individu‑ als with AD pathology of NFT stages I, III, and V. We stained and classified 25 brain regions in each using the anti‑ phospho‑tau monoclonal antibody AT8. We measured tau seeding in each of the 500 samples using a cell‑based tau “biosensor” assay in which induction of intracellular tau aggregation is mediated by exogenous tau assemblies. We observed a progressive increase in tau seeding according to NFT stage. Seeding frequently preceded NFT pathology, e.g., in the basolateral subnucleus of the amygdala and the substantia nigra, pars compacta. We observed seed‑ ing in brain regions not previously known to develop tau pathology, e.g., the globus pallidus and internal capsule, where AT8 staining revealed mainly axonal accumulation of tau. AT8 staining in brain regions identified because of tau seeding also revealed pathology in a previously undescribed cell type: Bergmann glia of the cerebellar cortex. We also detected tau seeding in brain regions not previously examined, e.g., the intermediate reticular zone, dorsal raphe nucleus, amygdala, basal nucleus of Meynert, and olfactory bulb. In conclusion, tau histopathology and seeding are complementary analytical tools. Tau seeding assays reveal pathology in the absence of AT8 signal in some instances, and previously unrecognized sites of tau deposition. The variation in sites of seeding between individuals could underlie differences in the clinical presentation and course of AD. Keywords: Alzheimer’s disease, AT8, FRET biosensor, Neurofibrillary tangles, Prion propagation, Tau seeding, NFT staging Introduction Tauopathies are a heterogeneous group of neurodegen- erative diseases defined by progressive brain accumula - tion of tau aggregates [35]. Sporadic Alzheimer’s disease (AD) is the most common, and is uniquely defined by *Correspondence: Marc.diamond@UTSouthwestern.edu Barbara E. Stopschinski and Kelly Del Tredici have equally contributed to coexistent tau and amyloid β pathology. AD neuropa- this work. thology includes intraneuronal somatic and axonal pre- Center for Alzheimer’s and Neurodegenerative Diseases, Peter O’Donnell tangles and neurofibrillary tangles (NFTs), neuropil Jr. Brain Institute, NL10.120, University of Texas Southwestern Medical Center, 6000 Harry Hines Blvd., Dallas, TX 75390, USA threads (NTs), extraneuronal ghost tangles, and amyloid Full list of author information is available at the end of the article β plaques. Tau pathology progresses in a defined and Talitha Louise Thomas, BS (July 7, 1982 – October 28, 2020) © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 2 of 19 characteristic pattern, allowing AD classification into or insoluble tau protein could be observed [25]. Further, different stages that correlate with antemortem clinical in fresh frozen brain tissue from individuals with AD, presentation [4]. the assay also detected seeding prior to neuropathologi- Aggregated tau protein is often phosphorylated [35], cal changes [14, 25]. We subsequently refined the method and the anti-phospho-tau monoclonal antibody AT8 to quantify seeding in fixed brain sections, which was [26] is typically used for detection and staging. AT8 equally reliable [30]. In fixed tissues from multiple AD binds phospho-serine 202 and phospho-threonine 205 patients at different NFT stages, we observed tau seed - on aggregated tau protein, and marks AD intraneu- ing first in the TRE and EC rather than in the locus coer - ronal pathology (pretangles and NFTs) [38]. AT8 signal uleus (LC), as we had previously hypothesized based on increases with disease progression and allows the defi - AT8 histopathology [28]. All prior analyses have been nition of NFT stages [3, 6]. In NFT stage I, AT8 marks confined to regions known to contain NFT pathology in selected brainstem nuclei and the transentorhinal cortex AD. The extent of tau seeding across widespread brain (TRE). In NFT stage II, AT8 marks the entorhinal cor- regions, however, is unknown. In this work we have used tex (EC) in the parahippocampal gyrus. In NFT stage an optimized biosensor cell line (TauRD(P301S)v2H) III, AT8 marks the CA1 region of the hippocampus, and [23] and we have tested for seeding across a large cross- neocortical regions of the temporal neocortex adjacent section of brain regions with and without known NFT to the TRE. In NFT stages IV and V, AT8 marks neo- pathology. We have generated a map of AD brain across cortical regions including the superior temporal gyrus NFT stages. This has revealed surprising patterns, and a (STG), and in NFT stage VI it marks primary neocorti- new type of cellular tau pathology. cal areas such as the visual cortex (VC). At NFT stages III to IV, more than 50% of individuals have signs of mild cognitive impairment, whereas at NFT stages V and VI Methods more than 90% of individuals exhibit signs of moderate Generation of biosensor cell line (TauRD(P301S)v2H) to severe dementia [26]. The severity of AD dementia A second generation of high sensitivity biosensor cells correlates with the extent of postmortem tau pathology termed v2H has recently been produced [23]. Using the [41, 56]. Additionally, longitudinal tau PET imaging has previously described lentiviral FM5-YFP plasmid [47], we confirmed the progression of tau pathology along NFT inserted the tau segment 246 to 378 with the P301S stages, and its correlation with neuronal dysfunction and mutation, replaced the human ubiquitin C (Ubc) pro- neurodegeneration [24, 27, 36, 44]. moter with a human cytomegalovirus (CMV) promoter, Progressive tau aggregation in AD occurs in patterns and replaced the YFP sequence with an mCerulean3 or consistent with neural networks [4]. Recent data from mClover3 coding sequence. To reduce translation read in vitro and in vivo experimental systems is consistent through of the tau ATG start site and increase transgene with trans-neuronal spread of pathology similar to prion expression, the sequence upstream of tau was modified disease, in which pathological species move from cell to encode an optimal Kozak sequence (5’-GCC ACC to cell, serving as templates to convert native tau into a ACC ATG GCC-3’). The GCC after the ATG start codon pathogenic aggregation-prone form, and thereby propa- encodes the amino acid A246 in tau. The sequence link - gating tau pathology among connected brain regions ing the tau segment and the coding sequence of the fluo - [10, 11, 47, 48]. It is unknown whether this mechanism rophore (mCerulean3 or mClover3) was optimized to the underlies progression in humans, however the presence following sequence: 5’- GSAGSAAGSGEF-3’. of soluble, non-aggregated pretangle pathogenic tau To create the v2H line, low passage HEK293T cells (P5) “seeds” in human brain that anticipate the development were thawed and passaged with antibiotic free media of NFT pathology is very consistent with this idea [20, twice before co-administration of P301S 246–378 tau- 28]. mCerulean3 tau-mClover3 lentivirus. After four pas- To detect tau seeding in biological samples, we previ- sages, single cells were isolated via fluorescence activated ously developed a sensitive and specific cell-based “bio - cell sorting (FACS) based on low, intermediate, and high sensor” seeding assay, in which the tau repeat domain brightness levels for both mCerulean3 and mClover3. containing a single disease-associated mutation (P301S) Monoclonal colonies were cultured to high cell number is fused to complementary fluorescent proteins (e.g., and tested by seeding assays with recombinant fibrils and cyan/yellow; cerulean/clover; clover/ruby), and expressed AD lysate. The v2H line was chosen for low background in cells of choice. The fusion proteins aggregate upon signal and high sensitivity, and used in subsequent seed- exposure to tau seeds, which is quantified by flow cytom - ing experiments as a next-generation biosensor based on etry [19, 25]. In a transgenic mouse model, the seeding previously established protocols [25]. assay scored positive many months before histopathology S topschinski et al. acta neuropathol commun (2021) 9:164 Page 3 of 19 Culture of biosensor cells Human autopsy samples Stable monoclonal v2H FRET biosensor cells were Human autopsy tissue used for this study was obtained grown in complete media: Dulbecco’s Modified Eagle’s from n = 20 individuals (10 females, 10 males, age range Medium (DMEM) (Gibco) with 10% fetal bovine serum 50–93 years, Table 1) and 1 control (1 female, 30 years (Sigma), 1% penicillin/streptomycin (Gibco) and 1% of age) in compliance with ethics committee guidelines Glutamax (Gibco). Cells were cultured and passaged at the University of Ulm as well as German federal and at 37 °C, 5% CO , in a humidified incubator. Dulbecco’s state law governing human tissue usage. Informed writ- phosphate buffered saline (Life Technologies) was used ten consent for autopsy was obtained previously from for washing the cells prior to harvesting with 0.05% the patients or their next of kin. Brains were fixed Trypsin–EDTA (Life Technologies). in a 4% buffered aqueous solution of formaldehyde for 14 days. Tissue blocks from 25 brain regions were excised and embedded in polyethylene glycol (PEG Mouse breeding for positive and negative controls 1000, Merck, Carl Roth Ltd, Karlsruhe, Germany). All experiments involving animals were approved 100 μm serial sections were collected and stained free- by the University of Texas Southwestern Medical floating, as described previously [3 ] (Table 2). Brain tis- Center Institutional Animal Care and Use Committee sue and the remaining tissue sections were stored for (IACUC). All mice were housed under a 12 h light/dark subsequent use in a 4% aqueous solution of formalde- cycle, and were provided food and water ad libitum. We hyde at 8–15 °C for up to 26 years. used tau KO mice containing a GFP-encoding cDNA integrated into exon 1 of the MAPT gene as a negative control. These mice were obtained from Jackson Labo - APOE genotyping ratory and maintained on a C57BL/6 J background. Apolipoprotein E status was available for 16/20 of the As a positive control, we obtained transgenic mice individuals studied (Table 1). The ε4 allele is a major expressing 1N4R P301S human tau under the murine genetic risk factor for sporadic AD [12], TDP-43 pro- prion promoter [57] from Jackson Laboratory, and teinopathy [55] and for dementia with Lewy bodies (DLB) maintained them on a B6/C3 background. The posi - and Parkinson’s disease dementia [9, 50, 53]. APOE geno- tive control mice were anesthetized at age 2.5 months typing was performed (E.G.) using a semi-nested poly- with isoflurane and kept at 37 °C throughout the inocu - merase chain reaction assay and restriction isotyping lation. We used 10 μL gas-tight Hamilton syringes to with restriction enzyme HhaI [21]. Genomic DNA was inject 10 µg of clone 9 cell protein lysate (previously extracted from formaldehyde-fixed and paraffin-embed - described in [29, 47]) in the left hippocampus (bregma: ded brain specimens using the manufacturer’s protocols -2.5 mm posterior, -2 mm lateral, -1.8 mm ventral). The (QIAamp DNA Mini Kit, Qiagen, Hilden, Germany). mice were euthanized 4 weeks later as described below for seeding experiments. Neuropathological staging Neuropathological staging and disease classification of Mouse sample collection and preparation AD-associated pathology were performed (H.B., K.D.T.) The mice were anesthetized with isoflurane and per - according to a previously published modified Gallyas fused with chilled PBS + 0.03% heparin. Brains were silver-iodide staining protocol [3, 4] for recognition of post-fixed in 4% PFA overnight at 4 °C and placed in phosphorylated somatic argyrophilic (fibrillary) neuropil 30% sucrose in PBS until further use. Brains were sec- threads (NTs) [1, 5] and neurofibrillary tangles (NFTs), as tioned at 50 μm with a freezing microtome and placed well as of extraneuronal ghost tangles (‘tombstone’ tan- into cryoprotectant (32% ethylene glycol, 16% w/v gles) that display weak staining with the Gallyas method sucrose, in 50 mM phosphate buffer pH 7.4, stored at and strong staining with the Campbell-Switzer silver-pyr- -20 °C). 1 mm punches were then isolated from the left idine method. In addition, AT8 immunohistochemistry hippocampus using Miltex disposable punch biopsy (IHC, monoclonal anti-PHF-Tau antibody, 1:2000; Clone tools. Four 4 mm punches were placed into 100 µl AT8; Pierce Biotechnology [Thermo Scientific] Waltham, EDTA buffer (1 mM EDTA, pH 8.0), heated for 25 min MA [38], was performed. In contrast to the Gallyas at 95 °C and allowed to cool down for ~ 15 min. The method, AT8 IHC visualizes the broadest spectrum of samples were then sonicated with a water bath sonica- intraneuronal pathological tau: argyrophilic NFTs of the tor (Qsonica Q700MPX with chiller and tubing set) at Alzheimer type, NTs in dendritic processes, and non- 4 °C at 50 amplitude for 60 min, and stored at -80 °C argyrophilic axonal aggregates and pretangles. AT8 IHC until further use. Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 4 of 19 Table 1 Demographic and neuropathological data Case f/m age brain wt NFT Aβ α-syn TDP-43 APOE Diagnoses 1 f 55 1455 I 0 0 − ε3/4 intraventricular hemorrhage 2 m 55 1780 I 0 0 − ε3/3 bronchopneumonia 3 f 50 1293 I 0 0 − ε3/3 PCOM aneurysm, SAB 4 m 50 1600 I 0 0 − ε4/4 M. Werlhof, hepatitis C 5 f 72 1070 I 3 0 − ε3/4 cardiac failure 6 f 79 1165 I 0 0 − ε3/3 coronary artery disease 7 f 93 1090 III (IV ) 2 0 − na craniocerebral trauma 8 m 68 1380 III 0 0 − na aspiration pneumonia 9 m 72 1500 III 0 0 − ε3/4 cardiac failure 10 m 74 1465 III 0 0 − ε4/4 malignant neoplasm 11 m 76 1520 III 4 0 + na pneumonia 12 f 81 1130 III 2 0 − ε3/3 acute myeloid leukemia 13 f 88 1495 III 3 0 − ε3/4 cerebral hemorrhage parietooccipital 14 m 84 1170 V ( VI) 3 0 + ε3/4 AD 15 f 84 1175 V 3 2 + ε3/3 AD, ILBD 16 m 58 1335 V 2 0 − ε3/4 AD, myocardial infarction 17 f 72 1185 V 4 0 − ε3/3 AD, craniocerebral trauma after falling 18 f 76 1405 V 4 0 − ε3/4 AD, craniocerebral trauma after falling 19 m 76 1205 V 5 0 − na AD, cardiac failure 20 m 78 1460 V 5 0 − ε3/4 AD Control 21 f 30 1315 0 0 0 − ε3/3 malignant neoplasm 20 cases fall into three NFT groups: NFT stage I (4 females, 2 males, 50–79 years); NFT stage III (3 females, 4 males, 68–93 years); NFT stage V (3 females, 4 males, 58–84 years). Abbreviations: f, m—female, male; age—age in years; brain wt—fresh brain weight in grams; NFT—Alzheimer’s disease-related neurofibrillary tangle stage using Gallyas silver-iodide staining; Aβ—amyloid-β deposition phase using 4G8 IHC; α-syn—Parkinson disease-related neuropathological stage using α-synuclein IHC; TDP-43—43-kDa TAR DNA-binding protein neuronal inclusions; APOE—APOE allele status; n/a -not available; AD—Alzheimer’s disease; ILBD— incidental Lewy body disease; PCOM -posterior communicating artery; SAB—subarachnoid bleeding detects ghost tangles less effectively than Gallyas silver polyclonal rabbit antibody recognizing the N-terminal of staining, or not at all. The character and relative merits of normal TDP-43 (1:5000; Proteintech, Manchester, UK) thioflavin-S staining, Gallyas and Campbell-Switzer silver [52]. We staged all cases for sporadic Parkinson’s dis- staining, as well as more conventional silver methods (the ease (PD), as described elsewhere [8] (Table 1). One case modified Bielschowsky and the Bodian methods) in rela - showed incidental α-synuclein-positive Lewy pathol- tion to tau isoforms and to IHC have been discussed in ogy; three cases displayed coincident TDP-43 immu- detail elsewhere [51, 54]. We evaluated the presence of noreactivity [17, 34] (Table 1). Histological slides were aging-related tau astrogliopathy (ARTAG) as proposed viewed with an Olympus BX61 microscope (Olympus by Kovacs [32]. We visualized and staged Aβ deposition Optical, Tokyo, Japan). Pathology was assessed semi- using the monoclonal anti-Aβ antibody 4G8 (1:5000; quantitatively on a four-point scale: 0 = no detect- Clone 4G8; BioLegend, San Diego, CA) as recommended able tau inclusions, + = mild (at least one or two previously [26]. Clinical AD classification included cases AT8-positive cell soma/somata); + + = moderate inclu- with tau stages III/V and Aβ phases ≥ 2 [13, 15] (Table 1). sions; + + + = severe inclusions. Digital micrographs We excluded other non-AD tauopathies, including of IHC-stained sections (Figs. 4, 5) were taken with an argyrophilic grain disease, progressive supranuclear Olympus XC50 camera (H.B.) using the Cell D Imaging palsy, Pick’s disease, corticobasal degeneration, and Software (Olympus, Münster, Germany). The extended Niemann-Pick disease type C. Separate sets of 100 µm focal imaging (EFI) function was used for stacking free-floating sections from all cases were immunostained images at different optical planes (Cell D Imaging Soft - using the following primary antibodies: (1) a monoclo- ware, Olympus, Münster, Germany). The EFI algorithm nal anti-syn-1 antibody (1:2000; Clone number 42; BD extracts the image features with the sharpest contrast Biosciences, Eysins, Switzerland) for detection of Lewy from all layers of the stack and merges them into a single body disease-related α-synuclein inclusions [26]; (2) a image. S topschinski et al. acta neuropathol commun (2021) 9:164 Page 5 of 19 Table 2 Brain regions sampled punch tool from positive and negative control mice (S.E.). Regions 1 Transentorhinal cortex ( TRE) Human sample preparation 2 Entorhinal cortex (EC, Brodmann Area 28) One 4 mm punch or two 3 mm punches were placed 3 Ammon’s horn, sector 1 (CA1, hippocampal formation) into 100 µl EDTA buffer (1 mM EDTA, pH 8.0), heated 4 Amygdala, basolateral subnucleus (AMY ) for 25 min at 95 °C and allowed to cool down at 4 °C 5 Superior (first) temporal gyrus (STG, Brodmann Area 22) for ~ 15 min. The samples were then sonicated with a 6 Transverse temporal gyrus of Heschl ( TTG, Brodmann Area 41) water bath sonicator (Qsonica Q700MPX with chiller 7 Primary visual neocortex (PV, Brodmann Area 17) and tubing set) at 4 °C at 50 amplitude for 60 min, and 8 Peristriate neocortex, high order sensory neocortex (Brodmann stored at −80 °C until further use. Area 19) 9 Anterior cingulate cortex—skeletomotor/emotion‑autonomic integration (ACC, Brodmann Areas 24/32) Transduction of biosensor cell lines, flow cytometry 10 Retrosplenial/posterior cingulate cortex—memory/visuospatial and seeding analysis orientation (RSC/PCC, Brodmann Areas 23/29/30) The seeding assay was conducted as previously 11 Putamen (PUT ) described with the following changes: biosensor cells 12 Globus pallidus (GP) were plated at a density of 25,000 cells/well in a 96-well 13 Mediodorsal complex of thalamus (MD) plate in a media volume of 130 µl per well. The mouse 14 Orbitofrontal cortex (OFC, Brodmann Area 11) and human tissue samples were thawed on ice, fol- 15 Substantia nigra, pars compacta (SNpc) lowed by thorough vortexing and incubation with Lipo- 16 Locus coeruleus (LC) fectamine 2000 for 30 min. 1 µl of tissue lysate with 17 Basal nucleus of Meynert(BN) 0.5 µl of lipofectamine and 18.5 µl of OptiMEM (Gibco, 18 Pontine gray (PG) Life Technologies) was added to each well, resulting 19 Inferior olivary nucleus (IO) in 20 µl total. For each experiment, negative controls 20 Cerebellar cortex (CC) received either Lipofectamine in OptiMEM (lipo- 21 Cerebellar dentate nucleus (DN) fectamine controls), or OptiMEM (buffer controls). 22 Internal capsule, anterior limb (IC) The lysate-lipofectamine mix was applied to the cells, 23 Terminal stria ( TS) and cells were incubated for an additional 72 h. Cells 24 Olfactory bulb (OB) were harvested with 0.05% trypsin and fixed in 2% PFA 25 Optic chiasm/tract (OC) for 10 min, then resuspended in flow cytometry buffer Punch biopsies were made from unstained sections of the 25 brain regions (HBSS plus 1% FBS and 1 mM EDTA). An LSRFortessa shown above using a 4 mm (3 mm for regions 19 and 22 in Experiment II) punch biopsy tool. Cross-contamination of seeding activity between individuals and SORP (BD Biosciences) was used to perform FRET regions was prevented by disposing the biopsy tool after each punch flow cytometry. We quantified FRET as previously described with the following modification: we identi - fied single cells that were double-positive for mCeru - Punch samples lean and mClover and subsequently quantified FRET From each case, including the negative human control, positive cells within this population. For each data punch samples were collected (K.D.T.) free-floating set, 3 technical replicates were included. Data analy- from unstained sections of the 25 brain regions shown sis was performed using FlowJo v10 software (Treestar in Table 2 with a punch biopsy tool (Kai Industries Co, Inc.), GraphPad Prism v8.4.3 for Mac OS X, and Excel Ltd. Japan) – with diameter of either 4 mm (result- 2 v16.16.25 (Microsoft). ing in a punch volume of ~ 1.257 mm ) or 3 mm (with estimated punch volume of ~ 0.706 mm ). The 3 mm punch device was only used for the internal capsule Statistical analyses (IC) and the inferior olivary nucleus (IO) in the 2nd Samples were collected at the University of Ulm and set of punches to ensure that the punches were con- cases were blinded prior to seeding analyses by B.E.S. at fined to the immediate target regions. To avoid cross UT Southwestern Medical Center. Flow cytometry gat- contamination of seeding between individuals and ing and analysis of seeding were completed prior to the regions, punch tools were used only once. Samples decoding and interpretation of the seeding results. All were encoded and all subsequent preparation and seed- statistical analysis was performed using GraphPad Prism ing assays were performed in a blinded fashion. Tissue v8.4.3 for Mac OS X and Excel v16.16.25 (Microsoft). punches were stored in 1 × TBS at 4 °C until use. Brain Statistical significance between seeding at different NFT tissue was collected in the same way with the 4 mm Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 6 of 19 stages was determined by performing a non-parametric the seeding assay was appropriately detecting tau seed- rank order test (Whitney-Mann test). Correlation analy- ing only when present and not detecting seeding when sis (linear regression) was performed and Spearman r tau was absent in human tissue. Therefore, we did not correlation was calculated to test the reproducibility of sample each of the 25 brain regions from case 21. We seeding between different experimental runs. used the average of all negative samples plus 3 standard deviations to determine the “positive” seeding thresh- old. We considered samples with seeding of 0.35% Generation of 3D tau seeding map and above to be positive, whereas samples below this Magnetic resonance imaging of the brain was performed threshold were considered negative. With a 3 SD cut- in a healthy volunteer using a 3 T Siemens Prisma off, we increased the specificity of our analysis at the MRI scanner, including acquisition of 3D T1-weighted cost of sensitivity. The coefficient of variation (defined MPRAGE sequence with 1 mm isotropic resolution. as standard deviation divided by the FRET average) was Segmentation of regions of interest (ROIs) within the used as a measure of precision and is high for samples brain was performed semi-automatically using Free- below the seeding threshold and low for samples above Surfer (version 5.3.0, http:// surfer. nmr. mgh. harva rd. edu). the threshold (Fig. 1). Given the relatively high speci- In cases where ROIs did not already exist in the Free- ficity of the assay, negative results do not rule out tau Surfer library, manual segmentation was performed by seeding in a given sample. an experienced board-certified neuroradiologist (F.F.Y.) using the Segmentation Editor tool in 3D Slicer (version 4.10.2, https:// www. slicer. org/). The weighted average tau Reproducibility between experimental runs and samples deposition from all subjects within each NFT stage group To test seeding within regions, we selected 50 punches was then applied to each ROI using a customized script from 20 individuals and obtained a second set of punches in Matlab (version R2015b). Brain regions were included from the same section as the 1st set whenever possible. only if at least one subject within each NFT stage exhib- If not possible (because of limited amount of tissue), the ited seeding. The Build Surface function within Mango 2nd punch was obtained from an immediately adjacent (version 4.1, http:// ric. uthsc sa. edu/ mango/) was then section. We then performed 2 additional seeding experi- used to visualize the ROIs within a 3D projection of a ments: In experiment II, the 2nd sample set was tested control brain. (run 1). In experiment III, both sample sets were thawed a second time and tested for seeding (run 2). Thus, the Results 1st sample set was tested after one freeze thaw cycle in Sampling of 25 brain regions across 20 individuals experiment I (run 1), and after a second freeze–thaw Previous publications from our group and others have cycle in experiment III (run 2). In the same way, the 2nd studied seeding in a limited number of brain samples sample set was tested in experiment II (run 1) and experi- from AD patients using the biosensor system [14, 20, 25, ment III (run 2) after one versus two freeze–thaw cycles 28]. We chose 20 individuals with confirmed AD pathol - (Fig. 2). For both experiments II and III, the coefficient ogy for this analysis (Table 1). Given that the differences of variation above the previously defined seeding thresh - in tau tangle pathology between NFT stage I/II, III/IV, old of 0.35% was low (Fig. 2b and c). We then used linear and V/VI are subtle, we limited our study to NFT stages regression to compare the sample sets and experimental I, III and V. Furthermore, we used the punch device runs. Seeding correlated well between different experi - established by Kaufman et al. 2018 [28] for more precise mental runs (run 1 and 2) of the same samples with R sampling of the 25 brain regions of interest as opposed to in the range of 0.7 – 0.8 (Fig. 2d, e). The reproducibility sampling by dissecting large tissue pieces (Table 2). of the seeding data between different samples (1st versus 2nd sample set) had low reliability (Fig. 2f, g). Seeding threshold determination To determine the lower limit of detection, tissue lysate was transduced into v2H biosensor cells. We quanti- Progressive accumulation of seeding within individuals fied the percentage of FRET positive cells on the flow To examine the progression of seeding across all brain cytometer as a correlate of intracellular tau seeding, regions, we created a heat map with the seeding for each compared to negative control samples. Negative sam- individual and each brain region (Fig. 3a). We also plotted ples included human tau-negative brain tissue from seeding for each individual brain region for all 20 indi- case 21 (taken from the pons), brain tissue from tau viduals (Fig. 3b) and generated a 3D seeding map for each knockout mice, and wells treated with lipofectamine NFT stage (Additional file 2: Figure S6). In general, seed- or buffer only. Note that the samples from case 21 ing increased with higher NFT stages in all brain regions were included as internal assay control to ensure that S topschinski et al. acta neuropathol commun (2021) 9:164 Page 7 of 19 Fig. 1 Seeding profile of cases (Experiment I). Punch biopsies were taken from 25 brain regions in 20 individuals (NFT stages I, III and V ), homogenized, and transduced into v2H biosensor cells. Seeding was quantified by determining the percentage of FRET positive cells on a flow cytometer. Each sample was tested in biological/technical triplicate, and the average is reported. a Negative controls included cells that were treated with: (1) lipofectamine (+ buffer); (2) buffer only; (3) tau negative human brain tissue (from individual number 21 in Table 1; 4) brain tissue from tau knockout mice. The average seeding for each condition is shown as percentage of FRET positive cells ± standard deviation. We used the average of all negative samples and 3 × their respective standard deviations to determine the seeding threshold at 0.35% (in red). Only samples with seeding above 0.35% were scored positive. b The FRET average for each sample in experiment 1 was plotted on a log scale against the coefficient of variation (measure of assay precision defined as the standard deviation divided by the FRET average). The coefficient of variation is low for samples with seeding above the threshold of ~ 0.35%. For samples with seeding below this threshold, the coefficient of variation is significantly larger. c The FRET average for controls with standard deviation was plotted on a linear scale against the coefficient of variation. For lipofectamine and buffer controls, the average and standard deviation were derived from all 97 respectively 12 wells in this experiment. For tau negative human tissue and tau knockout mouse tissue, averages and standard deviation were calculated for each of 2 triplicates and plotted separately. Note that all controls ± standard deviation are below the seeding threshold of 0.35%. Color code: lipofectamine (yellow), buffer (pink), tau negative human brain tissue (blue), tau knockout mouse brain tissue (green), samples from individuals 1–20 (black) examined. For unexpected seeding results (Table 3), we (n = 4/6) had positive seeding in the TRE. In 2 cases performed AT8 staining on selected brain regions to test (cases 1 and 3), seeding in the TRE was below the for tau deposition (Additional file 1: Table S1). threshold of detection despite the presence of 9 and 1 AT8-positive neuron(s) respectively in this region Tau seeding starts in the transentorhinal cortex (Additional file 1 : Table S1 and Fig. 4a). We detected seeding in the transentorhinal cor- tex (TRE) in all individuals at NFT stages III and V (Fig. 3a). At NFT stage I, the majority of individuals Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 8 of 19 ab Fig. 2 Reproducibility of seeding data. a To determine variation in seeding by region, we randomly selected a subset of 50 samples from the 1st sample set (tested in Exp I and shown in Fig. 1) and gathered a second set of punches from these brain regions (2nd sample set). Exp II tested seeding in the 2nd sample set (50 samples total plus controls). In Experiment III, seeding of both sample sets was tested a second time (100 samples total plus controls). In summary, both 1st and 2nd sample sets were tested in 2 separate runs, with the second run following a freeze–thaw cycle (Run 1 and 2). b and c show the plots for coefficient of variation versus FRET average for samples and controls in experiments II and III. The graphs were generated in the same way as for Experiment I in Fig. 1b and c. All control samples fall below the seeding threshold of 0.35%. Color code: lipofectamine (yellow), buffer (pink), tau negative human brain tissue (blue), tau knockout mouse brain tissue (green), samples from individuals 1–20 (black). d and e show a correlation analysis (linear regression) to test the reproducibility of seeding for the same sample sets between different experimental runs. f and g show the correlation analysis to test the reproducibility between different punches from the same brain region. Note that control samples were not included in graphs d to g Early seeding in the entorhinal cortex, CA1, amygdala, seeding. The entorhinal cortex (EC) and CA1/hippocam - and locus coeruleus pal formation scored positive in n = 2/6 NFT stage I Four additional brains regions demonstrated early cases, and remained positive at all higher NFT stages S topschinski et al. acta neuropathol commun (2021) 9:164 Page 9 of 19 Fig. 3 Seeding in 20 individuals, 25 brain regions (derived from Experiment I). a Seeding data heat map: Data points below the seeding threshold of 0.35% are colored in gray. Data points equal and above the seeding threshold are shaded with a graded color scale ranging from yellow (low) to red (high). Seeding data from human control brain (tau negative) was included as a comparison. b Seeding data from 25 brain regions plotted as individual graphs and separated according to NFT stage. Individual symbols (dot, square, triangle) represent data from individuals at each NFT stage. Statistical significance was determined by performing a non‑parametric rank order test ( Whitney‑Mann test) to compare NFT I vs. III, III vs. V, and I vs. V. ns = non‑significant, *p < 0.05, **p < 0.01. Errors bars show SD (n = 14/14) (Fig. 3a). Seeding was also detected in n = 1/6 at later stages (NFT V and VI) [4]. AT8 staining in n = 5 NFT stage I individuals in the basolateral subnucleus of seeding-positive cases (e.g., at NFT stages I and III, Addi- the amygdala (AMY), and this region remained positive tional file 1: Table S1) revealed immunopositive SNpc in the vast majority of NFT III (n = 6/7) and all NFT V axons in all individuals examined (Fig. 5e). However, cases (n = 7/7) (Figs. 3a and 5g). The locus coeruleus (LC) seeding above threshold could only be detected in one exhibited seeding in n = 1/6 at NFT stage I, and at NFT case (case 12) that also displayed AT8-positive neurons stages III and V the LC was positive for seeding in all (Additional file 1: Table S1). individuals tested, 2 of whom displayed strong seeding at The globus pallidus (GP) does not typically show AD- NFT stage V (Figs. 3a and 5h, i). This is broadly consist - associated tau pathology [7]. However, n = 4/7 cases at ent with our prior work [28]. NFT stage III and all cases at NFT stage V demonstrated seeding (Fig. 3). AT8 staining in 9 individuals revealed Intermediate seeding in 9 brain regions AT8 positivity in axons but not nerve cell somata (Addi- The superior temporal gyrus (STG), the peristriate neo - tional file 1: Table S1). In two separate punches from case cortex (area 19, PS), and the terminal stria (TS) scored 7 (Fig. 4g and h), we found a single AT8-positive neuronal positive for seeding in some individuals (n = 3/7) at NFT body in the GP. Seeding-positive axons in the GP could stage III, and in all individuals at NFT stage V (Fig. 3a). have their origins in the basal nucleus of Meynert (BN), The anterior cingulate cortex (ACC) demonstrated posi - which displayed seeding in n = 4/7 cases at NFT stage III tive seeding in most individuals at NFT stage III (n = 5/7) and in n = 7/7 NFT stage V cases (Fig. 3). and all NFT stage V cases, and the orbitofrontal cortex The olfactory bulb (OB) scored positive in n = 9/19 (OFC) as well as Meynert’s basal nucleus (BN) were seed- cases (Fig. 3a). We found no evidence of tau seeding at ing-positive in n = 4/6 cases at NFT stage III and in all NFT stage I (Fig. 3a). At NFT stage III, n = 4/7 individu- individuals at NFT stage V (Fig. 3a). als displayed seeding in this region, and at NFT stage V, We detected robust seeding in the substantia nigra, n = 5/6 individuals scored positive, 1 of which was par- pars compacta (SNpc) in most cases (n = 6/7) at NFT ticularly pronounced (case 15). No OB tissue from case stage III and n = 7/7 at NFT stage V (Fig. 3a). By contrast, 16 was available. tangle pathology is typically observed in this region only Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 10 of 19 Fig. 3 continued S topschinski et al. acta neuropathol commun (2021) 9:164 Page 11 of 19 Late seeding in 5 brain regions because of its close proximity to the intermediate reticu- Seeding at late NFT stages was present in 5 brain regions. lar zone (IRZ) (Figs. 4d, e and 5a-d), which can contain Heschl’s gyrus (TTG), the putamen (PUT), retrosplenial/ AT8-positive axons and/or neurons in early AD stages posterior cingulate cortex (RSC/PCC) and mediodorsal [46]. While we obtained mostly clean punches from IO, complex of thalamus (MD) were positive for seeding in we cannot exclude that a few contained edges of the only n = 1/7 to n = 2/7 individuals at NFT stage III, and IRZ, as noted in Additional file 1: Table S1 and shown in in most or all individuals at NFT stage V (Fig. 3). The pri - Fig. 5a and b. Use of a 3 mm punch helped to minimize mary visual neocortex (PV) had seeding in only n = 1/7 such occurrences. individuals at NFT stage III, and at NFT stage V n = 4/7 individuals scored positive. Cerebellar dentate nucleus and cerebellar cortex In two individuals, the cerebellar dentate nucleus (DN) and/or cerebellar cortex (CC) had seeding (Fig. 3a). Inter- Occasional/inconsistent seeding in 6 brain regions estingly, AT8 staining for the two positive CC regions We observed seeding and AT8-positive tau pathology in (cases 9 and 12) revealed AT8-positivity in the Bergmann some individuals at higher NFT stages (III and V) in 6 glia (as thorn-shaped astrocytes) and in their astrocytic additional brain regions, which is a new finding in some processes (Fig. 4j and k) rather than in neurons, which is cases. a novel ARTAG finding. In the DN of 4 cases, nerve cells plus axons were AT8-positive (Fig. 4i). Pontine gray Three individuals (n = 1 at NFT stage III and n = 2 at NFT stage V) scored positive for seeding in the pon- Internal capsule, anterior limb tine gray (PG). Of note, for case 7 (with positive seeding We observed seeding in the anterior limb of the internal of 0.79 in the PG), the dorsal raphe nucleus (DRN) was capsule in n = 1/7 case at NFT stage III, and in n = 4/7 punched accidentally instead of PG for the seeding assay cases at NFT stage V (Fig. 3a). AT8 staining in 5 cases (Fig. 4c). AT8 staining in n = 8/20 cases showed (mostly revealed positive axonal signals. The anterior portion of mild) AT8 changes in neurons and neurites/axons of the the inner capsule contains both axons of the frontopon- PG in n = 4 cases, of which n = 1 (case 16) had seeding tine projection and axons that originate in the mediodor- (Fig. 4b). sal complex of thalamus (MD) [42]. Neither the IC nor the OFC or MD were above threshold for seeding in NFT stage I, but this shifted at NFT stage III, and OFC and Inferior olivary nucleus MD displayed strong seeding at NFT stage V. 4 individuals (2 at NFT stage III and 2 at NFT stage V) had seeding in the inferior olivary nucleus (IO). AT8 staining in 9 cases was negative in neurons. However, Optic chiasm/tract three cases had tau-positive astrocytes consistent with Only n = 1/8 individual at NFT stage V (case 20) showed aging-related astrogliopathology (ARTAG) (Fig. 4f ), and positive seeding in the optic chiasm/tract (OC) (Fig. 3a). 6 cases had AT8-positive axons (Fig. 4d and e). Note that Additional tissue from this brain region was not available obtaining punches from the IO is technically challenging (See figure on next page.) Fig. 4 Phospho‑tau Histopathology (Part I). a The TRE of case 1 (female, 55 years, NFT I, Table 1) displayed a few AT8‑immunopositive neurons, but these were below the threshold for tau seeding (Fig. 3). With the exception of this case and case 3 (female, 50 years, NFT I, Table 1), the TRE of all 18 remaining individuals showed some degree of tau seeding activity (0.30–17.90, Fig. 3). b In case 16 (male, 58 years, NFT V, Table 1), moderate AT8 pathology (neuronal somata, axons) and low tau seeding (1.07) were detectable in the PG. c Punches mistakenly located in the dorsal raphe nucleus (DRN) of one individual (case 7, female, 93 years, NFT III, Table 1) showed AT8‑immunoreactive neuronal pathology combined with mild tau seeding (0.79). d Some IO punches (case 7), with no AT8‑positive cell bodies, contained isolated AT8‑positive axons and low tau seeding activity (0.60). e Framed area in d at higher magnification. Arrows indicate two AT8‑positive axons. f In the IO of case 7 (same individual as in c-e and g, h) we noted unexpected and marked aging‑related tau astrogliopathy, ARTAG (example in the framed area). g, h A single AT8‑positive neuron was seen in this punch from the GP accompanied by numerous AT8‑immunoreactive axons (see also Fig. 5f ), the latter possibly originating in the basal nucleus of Meynert (BN). Tau seeding in the GP of this case was moderate (2.65); notably, the BN of this individual also featured moderate tau seeding (5.46). I. The DN of case 9 (male, 72 years, NFT III, Table 1) had few AT8‑positive cell bodies (pretangles) and was sub ‑threshold for tau seeding (0.25). In cases at NFT stages I and III, only one (case 12, female, 81 years, NFT III, Table 1) had mild tau seeding in this region (0.60). j, k Notably, in the CC of case 12 (female, 81 years, NFT III, Table 1), the Bergmann glia (here, as thorn‑shaped astrocytes) and, in the molecular layer, their astrocytic processes were AT8‑immunopositive and also had tau seeding (1.39). By contrast, the neurons in the Purkinje layer were AT8‑negative and negative for tau seeding. Tau seeding (1.74) was detected in the CC of one additional individual (case 9, male, 72 years, NFT III, Table 1, Fig. 3) Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 12 of 19 Fig. 4 (See legend on previous page.) S topschinski et al. acta neuropathol commun (2021) 9:164 Page 13 of 19 to obtain reliable AT8 staining. Seeding in the OC might especially in closely adjacent anatomical regions, such as originate in axons from the DRN, the lower raphe nuclei, the IO and IRZ. However, smaller diameter punches also LC, amygdala, EC, superior colliculus, and/or the retina increased the variance in tau seeding detected, especially [42, 49]. in areas with scarce tau pathology, e.g., the IC. Further, given the small size of some brain regions examined we occasionally sampled adjacent sections, which may have ApoE status does not influence seeding in our study contributed to the variability. Finally, the high threshold ApoE genotyping was available for n = 16/20 cases. 8 car- to register a positive signal may have contributed to the ried at least one ApoE ε4 allele, and 2 were homozygous variability in regions with low numbers of tau-positive for ApoE ε4 (Table 1). We detected no significant influ - neurons. In prior work [28], adjacent punches from the ence of ApoE on the seeding in all cases comparing E4 TRE/EC correlated relatively well, which may be due carriers versus non-carriers. to its more easily defined structure, or more homoge - neously distributed pathology than many of the brain regions studied here. Other tauopathies can be excluded in our cohort None of the cases displayed the tauopathy changes spe- cific for CTE [37] or PSP [16, 33, 43, 45]. Seeding increases with neuropathological stage Some brain regions studied are not known to develop Discussion tau pathology in AD, yet we observed tau seed- We used an ultrasensitive biosensor assay to create a ing across all 25. Many regions included in this study comprehensive map of seeding in the AD brain, analyz- have not previously been studied for tau seeding. Our ing 25 brain regions (15 of which had not been analyzed predictions regarding seeding were based on prior previously) across 20 individuals with AD pathology neuropathological data and our knowledge of the neu- at NFT stages I, III, and V. We used a punch biopsy for roanatomical connections that determine input and precise sampling of regions of interest, and defined the output for these brain regions. Our predictions were positive seeding at 3 SD above the average of all negative accurate in many cases (Additional file 1: Table S2), e.g., samples tested, a threshold with relatively high specificity as predicted, we found early seeding in the transen- and low sensitivity. torhinal and entorhinal cortex; seeding at intermedi- ate stages in the superior temporal gyrus (STG) and the anterior cingulate cortex (ACC); and seeding at Variability within brain regions intermediate/late stages in the putamen and the medi- We observed variability between different punches from odorsal complex of thalamus (MD). These findings were the same brain regions, highlighting the importance in consistent with the tau progression pathway described future studies of sampling larger brain volumes or aver- in prior studies [4, 14, 28]. aging multiple biopsies. We used small punches with a diameter of 3–4 mm to increase sampling precision, (See figure on next page.) Fig. 5 Phospho‑tau Histopathology (Part II). a A 4 mm punch from the IO of case 16 (male, 58 years, NFT V ) included portions of the immediately adjacent intermediate zone (IRZ) with strongly AT8‑positive neurons and axons. In the IO itself, a single AT8‑immunoreactive axon was detectable and some perivascular ARTAG was present, but none of the neurons there were AT8‑immunoreactive. Tau seeding was below threshold (0.24). In at least two additional individuals (cases 12 and 17), prominent AT8‑positive neurons and axons in the IRZ accompanied by IO AT8‑negative neurons as well as perivascular ARTAG may have accounted for the tau seeding signals in IO punches (2.29 and 4.74). b Framed area in a at higher magnification showing AT8‑positive tau pathology at the punch edge. c A 3 mm punch with edges free of AT8 pathology from case 15 (female, 84 years, NFT V ), in which no portions of the IRZ were included. d Framed area in c at higher magnification displaying a clean punch edge and, directly beyond it, AT8‑positive pathology in the IRZ. Tau seeding activity in the IO of this case was below threshold (0.08). e AT8‑positive axon (arrows) in the substantia nigra, pars compacta (SNpc) of case 13 (female, 88 years, NFT III). Some tau seeding in the SNpc was present in 13/20 individuals (0.81–14.20), and was exceeded in the lower brainstem only by 15/20 cases in the LC (1.00–19.43). f Arrows point to AT8‑positive axons in the GP of case 7 (female, 93 years, NFT III) (see also Fig. 4g, h). g The highest propensity for tau seeding (33.77) was detected in the basolateral subnucleus of the amygdala (AMY ) of case 18 (female, 76 years, Stage V ), where severe diffuse neuronal AT8‑pathology was accompanied by some AT8‑immunopositive astrocytes (ARTAG). h, i Tau seeding activity in the LC first became more pronounced during NFT stages III and V, e.g., 11.96 in case 7 (female, 93 years, NFT III) (h); 19.43 in case 19 (male, 76 years, NFT III); and 5.71 in case 10 (male, 74 years, NFT III) in micrograph i. Arrow in i points to extraneuronal neuromelanin lying free in the neuropil after severe neuronal loss Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 14 of 19 Fig. 5 (See legend on previous page.) Tau seeding starts in the TRE both individuals 1 and 3 contained up to 9 AT8-positive The majority of individuals tested in our study scored neurons and scored negative for seeding, possibly due positive in the TRE, excepting 2 individuals at NFT to the low density of AT8-positive neurons and the rela- stage 1 (cases 1 and 3). Interestingly, TRE sections of tively high seeding threshold we used (Additional file 1: S topschinski et al. acta neuropathol commun (2021) 9:164 Page 15 of 19 Table 3 Unexpected seeding results Brain region Expectation Reference for expectation Findings 1‑ TRE Seeding in all cases Kaufmann et al. [28] Unexpected = > n = 2/6 cases at NFT stage I did not show seeding 4‑AMY Seeding starting at late NFT stages Braak et al. [4] Unexpected = > seeding in n = 1/6 NFT stage I case, in n = 6/7 NFT stage III cases and in all NFT stage V cases 6‑ TTG Seeding starting at late NFT stages ( V‑ VI) Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case; expectedly in n = 6/7 NFT stage V cases 12‑ GP Seeding not expected since it typically does not Braak and Del Tredic [6] Unexpected = > seeding in n = 3/7 NFT stage develop tau pathology (in contrast to Aβ plaques) III cases and in n = 6/7 NFT stage V cases in AD 15‑SN Seeding starting at late NFT stages ( V‑ VI) Braak et al. [4] Unexpected = > seeding seen earlier—in most NFT stage III cases (n = 6/7) and in all NFT stage V cases 18‑PG Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case and in n = 2/7 NFT stage V cases Note: The dorsal raphe nucleus (DRN) in case 7 was punched accidentally and likely explains the seeding in this sample 19‑IO Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 2/7 NFT stage III cases and in n = 2/7 NFT stage V cases Note: Some punches from IO were contaminated by over‑ lapping AT8‑positive neurons/axons in the IRZ 20‑ CC Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 2/7 NFT stage III cases 21‑DN Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case and in n = 1/7 NFT stage V case 22‑IC Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case and in n = 4/7 NFT stage V cases 24‑ OB Seeding starting at early/intermediate NFT stages Attems et al. [2] As expected, seeding in n = 4/7 NFT stage III cases; (II‑III) however, unexpected = > in only n = 5/6 NFT stage V cases 25‑ OC Seeding expected DeVos et al. [14] Unexpected = > seeding in only one individual (case 20) Summary of seeding results that were unexpected based on seeding results and/or neuropathology data from prior studies. For a complete list of all brain regions and expectations prior to the study refer to Additional file 1: Table S2. For unexpected results, AT8 IHC was performed as listed in Additional file 1: Table S1 Table S1). In other words, the number of AT8-positive the seeding assay reliably detects monomeric and oligo- structures (i.e., “severity” of pathology) in the TRE may meric tau seeds [39, 40], and we speculate that the pres- not account for the level of seed-competent tau. Prior ence of seed-competent tau monomer and oligomers can work from our group indicates that early tau pathology precede the presence of AT8-positive tangles. Studies in starts in the TRE/EC region [28]. Of note, Kaufman et al. the PS19 mouse model support these conclusions [25]. [28] did not separate the TRE and EC, as we did here. The differences in seeding between TRE and EC were subtle: Tau seeding occurs in unanticipated brain regions compared to the EC, the TRE had seeding in more indi- We detected tau seeding in brain regions not known to viduals (n = 4 vs. n = 2) at NFT stage I, and the average develop tau pathology, i.e. the GP, IC, PG, IO, and cer- seeding was slightly higher at NFT stage I. However, the ebellum. The seeding in these brain regions was low slightly earlier involvement of TRE is consistent with its overall, and neuropathological findings/AT8 staining and classification as NFT stage I [4]. seeding did not necessarily correlate with each other. In some regions, subsequent AT8 staining revealed the pres- Tau seeding precedes tau pathology ence of tau predominantly in astrocytes (see below). We Some brain regions show seeding significantly ear - also detected tau seeding in regions not known to exhibit lier than predicted based on neuropathological staging tau seeds, although observed to develop neurofibrillary (SNpc, AMY, TTG/Heschl’s gyrus). This confirms prior pathology during AD. These included the IRZ, DRN, observations that tau seeding can precede the develop- AMY, BN, and OB. The seeding was not always consist - ment of tau pathology as detected by immunohistochem- ent between connected anatomical regions within each istry [14, 20]. Work from our laboratory has shown that Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 16 of 19 Summary and conclusion individual case, since punches from different regions In the future, we will focus on selected brain regions were taken randomly from the right and left hemisphere, with notable and/or new results, such as the IRZ, AMY, and we did not intend to analyze direct anatomical con- BN, IC, GP, cerebellum, and OB. Moreover, an analy- nectivity between specific brain punches. Therefore, we sis of tau strains in all of the regions studied here and cannot speculate about anatomical pathways that could in the hypothalamus [4] would be informative. Impor- explain seeding in these samples. However, we note that tantly, although our data are limited by the number of the AT8-positive axons observed in the GP could belong individuals analyzed (n = 6–7 per NFT stage), we have to the BN. Seeding levels in the BN in all such cases were observed tau seeding in regions where AT8 staining always comparatively higher than in their respective GP is negative. This suggests that tau pathology extends punches (Fig. 3a). beyond the brain regions that are routinely included in AD staging, and is highly variable between individu- Aging-related tau astrogliopathy (ARTAG) als. Thus, immunohistochemistry and seeding are two In some brain regions with unexpected presence of tau fundamentally different and yet complementary meth - seeding (globus pallidus, inferior olivary nucleus, cer- ods for assessing tau pathology in AD. It remains to be ebellar cortex), we detected tau pathology in astrocytes determined whether a combination of these metrics or in Bergmann glia. It is not known how abnormal tau will help explain the diversity of AD presentation. in astrocytes accumulates [18]. ARTAG develops mainly, but not exclusively, in individuals over 60 years of age. Supplementary Information The two major cytomorphologies are thorn-shaped astro - The online version contains supplementary material available at https:// doi. cytes (TSAs) and granular or fuzzy tau immunoreactiv- org/ 10. 1186/ s40478‑ 021‑ 01255‑x. ity in astrocytic processes (GFA). TSAs occur in subpial, subependymal, or perivascular areas, as well as white Additional file 1: Tables S1 and S2. See separate online file. matter [31, 32]. It is now generally accepted that astro- Additional file 2: Figure S6 (a). Tau seeding map of NFT stage I. See glia either have no endogenous tau [22] or, very low lev- separate online files. MR imaging of the brain was performed in a healthy els [31]. It is unclear why tau accumulates in astrocytes, volunteer. Segmentation of regions of interest (ROIs) within the brain and whether it transfers between them. We hypothesize was performed semi‑automatically or by a board‑ certified neuroradiolo ‑ gist (F.F.Y.). The weighted average tau deposition from all subjects within that they phagocytize it from the surrounding intersti- each NFT stage group was then applied to each ROI. Brain regions were tial space, while tau seeds might originate in neurons included only if at least one subject within each NFT stage had positive [18]. Ferrer et al. have described seeding in AT8-positive tau seeding results. The ROIs were then visualized within a 3D projection of the volunteer’s brain. For clarity, 3D representation is limited to one astrocytes in cases displaying minimal intraneuronal tau hemisphere. A graded color scale from white to yellow to orange to red pathology without determining where seed competent (the same color coding as used in Fig. 3) indicates increasing amount of tau originates [18]. The pathophysiological significance seeding. Brain regions not sampled in this study are colored in grey. Note that many brains regions that are traditionally not included in AD pathol‑ of ARTAG for AD is unknown [31] and will require fur- ogy staging exhibit tau seeding, some of them at early NFT stages. ther studies. Our prior studies of tau strains indicate that Additional file 3: Figure S6 (b). Tau seeding map of NFT stage III. See some preferentially involve astrocytes [29, 47]. separate online files. MR imaging of the brain was performed in a healthy volunteer. Segmentation of regions of interest (ROIs) within the brain was performed semi‑automatically or by a board‑ certified neuroradiolo ‑ Tau seeding in the cerebellum and tau pathology gist (F.F.Y.). The weighted average tau deposition from all subjects within each NFT stage group was then applied to each ROI. Brain regions were in Bergmann glia included only if at least one subject within each NFT stage had positive We predicted that the cerebellar dentate nucleus (DN) tau seeding results. The ROIs were then visualized within a 3D projection and the cerebellar cortex (CC) would not exhibit seeding, of the volunteer’s brain. For clarity, 3D representation is limited to one hemisphere. A graded color scale from white to yellow to orange to red however, for each region, n = 2/20 cases did. We did not (the same color coding as used in Fig. 3) indicates increasing amount of observe AT8-positive tau pathology in the somatoden- seeding. Brain regions not sampled in this study are colored in grey. Note dritic or axonal compartments of cerebellar neurons, e.g., that many brains regions that are traditionally not included in AD pathol‑ ogy staging exhibit tau seeding, some of them at early NFT stages. granule cells, Purkinje cells, or stellate/basket cells. Nota- Additional file 4: Figure S6 (c). Tau seeding map of NFT stage V. See bly, AT8 staining for the 2 seeding-positive CC regions separate online files. MR imaging of the brain was performed in a healthy revealed AT8 staining of Bergmann glia rather than volunteer. Segmentation of regions of interest (ROIs) within the brain neurons and axons (Fig. 4j, k) (see remarks under the was performed semi‑automatically or by a board‑ certified neuroradiolo ‑ gist (F.F.Y.). The weighted average tau deposition from all subjects within previous heading ARTAG). We previously observed low each NFT stage group was then applied to each ROI. Brain regions were tau seeding in the cerebellum at late stages [20] but tau included only if at least one subject within each NFT stage had positive pathology in Bergmann glia has not been described. The tau seeding results. The ROIs were then visualized within a 3D projection of the volunteer’s brain. For clarity, 3D representation is limited to one findings raise the possibility of a sub-type of AD, with a hemisphere. A graded color scale from white to yellow to orange to red tau strain that preferentially affects Bergmann glia cells. (the same color coding as used in Fig. 3) indicates increasing amount of S topschinski et al. acta neuropathol commun (2021) 9:164 Page 17 of 19 4. Braak H, Braak E (1991) Neuropathological stageing of Alzheimer‑related seeding. Brain regions not sampled in this study are colored in grey. Note changes. Acta Neuropathol 82:239–259. https:// doi. org/ 10. 1007/ BF003 that many brains regions that are traditionally not included in AD pathol‑ ogy staging exhibit tau seeding, some of them at early NFT stages. 5. Braak H, Braak E, Grundke‑Iqbal I, Iqbal K (1986) Occurrence of neuropil threads in the senile human brain and in Alzheimer’s disease: a third location of paired helical filaments outside of neurofibrillary tangles and Acknowledgements neuritic plaques. Neurosci Lett 65:351–355. https:// doi. org/ 10. 1016/ 0304‑ We acknowledge support from the Aging Minds Foundation (B.E.S., M.I.D.), the 3940(86) 90288‑0 Cure Alzheimer’s Foundation (B.E.S., M.I.D.), the Berry Cox Foundation (B.E.S, 6. Braak H, Del Trecidi K (2015) Neuroanatomy and pathology of sporadic M.I.D.), the King Foundation (B.E.S.), the Chan‑Zuckerberg Initiative (M.I.D.), the Alzheimer’s disease. Adv Anat Embryol Cell Biol 215:1–162 NIH/NIA RF1AG059689 (M.I.D.), the Hans & Ilse Breuer Foundation, Frank‑ 7. Braak H, Del Tredici K (2014) Are cases with tau pathology occurring furt am Main, Germany (H.B., K.D.T ), Ms. Simone Feldengut (silver staining, in the absence of Abeta deposits part of the AD‑related pathologi‑ immunohistochemistry), and Mr. David Ewert (Figure 4, 5 layouts) for skillful cal process? Acta Neuropathol 128:767–772. https:// doi. org/ 10. 1007/ technical assistance. s00401‑ 014‑ 1356‑1 8. Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN, Braak E (2003) Authors’ contributions Staging of brain pathology related to sporadic Parkinson’s disease. Neuro‑ B.E.S designed and performed all cell culture and flow cytometry experiments. biol Aging 24:197–211. https:// doi. org/ 10. 1016/ s0197‑ 4580(02) 00065‑9 S.E. assisted with animal tissue collection and preparation of animal tissue. 9. Bras J, Guerreiro R, Darwent L, Parkkinen L, Ansorge O, Escott‑Price V, K.D.T. and H.B. performed all human tissue collection, IHC and neuropatho‑ Hernandez DG, Nalls MA, Clark LN, Honig LS et al (2014) Genetic analysis logical staging. E.G. performed APOE genotyping. F.F.Y. generated the 3D tau implicates APOE, SNCA and suggests lysosomal dysfunction in the seeding maps. M.I.D. assisted with the design and interpretation of all flow etiology of dementia with Lewy bodies. Hum Mol Genet 23:6139–6146. cytometry experiments. All authors assisted in the writing and figure prepara‑ https:// doi. org/ 10. 1093/ hmg/ ddu334 tion for this manuscript. All authors read and approved the final manuscript. 10. Calafate S, Buist A, Miskiewicz K, Vijayan V, Daneels G, de Strooper B, de Wit J, Verstreken P, Moechars D (2015) Synaptic contacts enhance cell‑to ‑ Availability of data and materials cell tau pathology propagation. Cell Rep 11:1176–1183. https:// doi. org/ The datasets used and/or analyzed during the current study are available from 10. 1016/j. celrep. 2015. 04. 043 the corresponding author by reasonable request. 11. Clavaguera F, Goedert M, Tolnay M (2010) Induction and spreading of tau pathology in a mouse model of Alzheimer’s disease. Med Sci (Paris) 26:121–124. https:// doi. org/ 10. 1051/ medsci/ 20102 62121 Declarations 12. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak‑ Vance MA (1993) Gene dose of apoli‑ Ethics approval and consent to participate poprotein E type 4 allele and the risk of Alzheimer’s disease in late onset All animal maintenance and experiments adhered to the animal care and use families. Science 261:921–923. https:// doi. org/ 10. 1126/ scien ce. 83464 43 protocols of the University of Texas Southwestern Medical Center. The autopsy 13. Crary JF, Trojanowski JQ, Schneider JA, Abisambra JF, Abner EL, brains used for this study were obtained in compliance with Ulm University Alafuzoff I, Arnold SE, Attems J, Beach TG, Bigio EH et al (2014) Primary ethics committee guidelines as well as German federal and state law govern‑ age‑related tauopathy (PART ): a common pathology associated with ing human tissue usage. human aging. Acta Neuropathol 128:755–766. https:// doi. org/ 10. 1007/ s00401‑ 014‑ 1349‑0 Consent for publication 14. DeVos SL, Corjuc BT, Oakley DH, Nobuhara CK, Bannon RN, Chase A, Not applicable. Commins C, Gonzalez JA, Dooley PM, Frosch MP et al (2018) Synaptic tau seeding precedes tau pathology in human Alzheimer’s disease brain. Competing interests Front Neurosci 12:267. https:// doi. org/ 10. 3389/ fnins. 2018. 00267 MID is co‑ developer of an anti‑tau antibody currently in clinical trials (C2N 15. Duyckaerts C, Braak H, Brion JP, Buee L, Del Tredici K, Goedert M, Halliday 8E12 [NCT02494024]). The remaining authors declare that they have no G, Neumann M, Spillantini MG, Tolnay M et al (2015) PART is part of Alz‑ competing interests. heimer disease. Acta Neuropathol 129:749–756. https:// doi. org/ 10. 1007/ s00401‑ 015‑ 1390‑7 Author details 16. Ebashi M, Ito Y, Uematsu M, Nakamura A, Hirokawa K, Kamei S, Uchihara Center for Alzheimer’s and Neurodegenerative Diseases, Peter O’Donnell Jr. T (2019) How to demix Alzheimer‑type and PSP ‑type tau lesions out of Brain Institute, NL10.120, University of Texas Southwestern Medical Center, their mixture ‑hybrid approach to dissect comorbidity. Acta Neuropathol 6000 Harry Hines Blvd., Dallas, TX 75390, USA. Clinical Neuroanatomy Section/ Commun 7:71. https:// doi. org/ 10. 1186/ s40478‑ 019‑ 0708‑4 Department of Neurology, Center for Biomedical Research, University of Ulm, 17. Elobeid A, Libard S, Leino M, Popova SN, Alafuzoff I (2016) Altered pro ‑ Ulm, Germany. Institute of Clinical Neuroanatomy, J. W. Goethe University, teins in the aging brain. J Neuropathol Exp Neurol 75:316–325. https:// Frankfurt am Main, Germany. Department for Radiology, Neuroradiology Divi‑ doi. org/ 10. 1093/ jnen/ nlw002 sion, University of Texas Southwestern Medical Center, Dallas, TX, USA. 18. Ferrer I, Garcia MA, Gonzalez IL, Lucena DD, Villalonga AR, Tech MC, Llor‑ ens F, Garcia‑Esparcia P, Martinez‑Maldonado A, Mendez MF et al (2018) Received: 3 September 2021 Accepted: 3 September 2021 Aging‑related tau astrogliopathy (ARTAG): not only tau phosphorylation in astrocytes. Brain Pathol 28:965–985. https:// doi. org/ 10. 1111/ bpa. 12593 19. Furman JL, Holmes BB, Diamond MI (2015) Sensitive Detection of Pro‑ teopathic Seeding Activity with FRET Flow Cytometry. J Vis Exp: e53205 https:// doi. org/ 10. 3791/ 53205 References 20. Furman JL, Vaquer‑Alicea J, White CL 3rd, Cairns NJ, Nelson PT, Diamond 1. Ashford JW, Soultanian NS, Zhang SX, Geddes JW (1998) Neuropil threads MI (2017) Widespread tau seeding activity at early Braak stages. Acta are collinear with MAP2 immunostaining in neuronal dendrites of Alz‑ Neuropathol 133:91–100. https:// doi. org/ 10. 1007/ s00401‑ 016‑ 1644‑z heimer brain. J Neuropathol Exp Neurol 57:972–978. https:// doi. org/ 10. 21. Ghebremedhin E, Braak H, Braak E, Sahm J (1998) Improved method 1097/ 00005 072‑ 19981 0000‑ 00009 facilitates reliable APOE genotyping of genomic DNA extracted from for‑ 2. Attems J, Lintner F, Jellinger KA (2005) Olfactory involvement in aging maldehyde‑fixed pathology specimens. J Neurosci Methods 79:229–231. and Alzheimer’s disease: an autopsy study. J Alzheimers Dis 7:149–157; https:// doi. org/ 10. 1016/ s0165‑ 0270(97) 00197‑0 discussion 173–180. https:// doi. org/ 10. 3233/ jad‑ 2005‑ 7208 22. Goedert M, Spillantini MG, Potier MC, Ulrich J, Crowther RA (1989) Clon‑ 3. Braak H, Alafuzoff I, Arzberger T, Kretzschmar H, Del Tredici K (2006) ing and sequencing of the cDNA encoding an isoform of microtubule‑ Staging of Alzheimer disease‑associated neurofibrillary pathology associated protein tau containing four tandem repeats: differential using paraffin sections and immunocytochemistry. Acta Neuropathol expression of tau protein mRNAs in human brain. EMBO J 8:393–399 112:389–404. https:// doi. org/ 10. 1007/ s00401‑ 006‑ 0127‑z Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 18 of 19 23. Hitt BD, Vaquer‑Alicea J, Manon VA, Beaver JD, Kashmer OM, Garcia JN, 40. Mirbaha H, Holmes BB, Sanders DW, Bieschke J, Diamond MI (2015) Tau Diamond MI (2021) Ultrasensitive tau biosensor cells detect no seeding trimers are the minimal propagation unit spontaneously internalized to in Alzheimer’s disease CSF. Acta Neuropathol Commun 9:99. https:// doi. seed intracellular aggregation. J Biol Chem 290:14893–14903. https:// doi. org/ 10. 1186/ s40478‑ 021‑ 01185‑8org/ 10. 1074/ jbc. M115. 652693 24. Hoenig MC, Bischof GN, Seemiller J, Hammes J, Kukolja J, Onur OA, Jessen 41. Nelson PT, Alafuzoff I, Bigio EH, Bouras C, Braak H, Cairns NJ, Castellani F, Fliessbach K, Neumaier B, Fink GR et al (2018) Networks of tau distribu‑ RJ, Crain BJ, Davies P, Del Tredici K et al (2012) Correlation of Alzheimer tion in Alzheimer’s disease. Brain 141:568–581. https:// doi. org/ 10. 1093/ disease neuropathologic changes with cognitive status: a review of the brain/ awx353 literature. J Neuropathol Exp Neurol 71:362–381. https:// doi. org/ 10. 1097/ 25. Holmes BB, Furman JL, Mahan TE, Yamasaki TR, Mirbaha H, Eades WC, NEN. 0b013 e3182 5018f7 Belaygorod L, Cairns NJ, Holtzman DM, Diamond MI (2014) Proteo‑ 42. Nieuwenhuys R, Voogd J, Huijzen C (2008) The human central nervous pathic tau seeding predicts tauopathy in vivo. Proc Natl Acad Sci U S A system: a synopsis and atlas, 4th edn. Springer, Berlin 111:E4376‑4385. https:// doi. org/ 10. 1073/ pnas. 14116 49111 43. Nogami A, Yamazaki M, Saito Y, Hatsuta H, Sakiyama Y, Takao M, Kimura 26. Hyman BT, Phelps CH, Beach TG, Bigio EH, Cairns NJ, Carrillo MC, Dickson K, Murayama S (2015) Early stage of progressive supranuclear palsy: a DW, Duyckaerts C, Frosch MP, Masliah E et al (2012) National institute on neuropathological study of 324 consecutive autopsy cases. J Nippon Aging‑Alzheimer’s association guidelines for the neuropathologic assess‑ Med Sch 82:266–273. https:// doi. org/ 10. 1272/ jnms. 82. 266 ment of Alzheimer’s disease. Alzheimers Dement 8:1–13. https:// doi. org/ 44. Ossenkoppele R, Schonhaut DR, Scholl M, Lockhart SN, Ayakta N, Baker 10. 1016/j. jalz. 2011. 10. 007 SL, O’Neil JP, Janabi M, Lazaris A, Cantwell A et al (2016) Tau PET patterns 27. Ishiki A, Okamura N, Furukawa K, Furumoto S, Harada R, Tomita N, Hiraoka mirror clinical and neuroanatomical variability in Alzheimer’s disease. K, Watanuki S, Ishikawa Y, Tago T et al (2015) Longitudinal assessment of Brain 139:1551–1567. https:// doi. org/ 10. 1093/ brain/ aww027 tau pathology in patients with Alzheimer’s disease using [18F]THK‑5117 45. Rüb U, Del Tredici K, Schultz C, de Vos RA, Jansen Steur EN, Arai K, Braak positron emission tomography. PLoS ONE 10:e0140311. https:// doi. org/ H (2002) Progressive supranuclear palsy: neuronal and glial cytoskeletal 10. 1371/ journ al. pone. 01403 11 pathology in the higher order processing autonomic nuclei of the lower 28. Kaufman SK, Del Tredici K, Thomas TL, Braak H, Diamond MI (2018) Tau brainstem. Neuropathol Appl Neurobiol 28:12–22. https:// doi. org/ 10. seeding activity begins in the transentorhinal/entorhinal regions and 1046/j. 0305‑ 1846. 2001. 00374.x anticipates phospho‑tau pathology in Alzheimer’s disease and PART. Acta 46. Rüb U, Del Tredici K, Schultz C, Thal DR, Braak E, Braak H (2001) The Neuropathol 136:57–67. https:// doi. org/ 10. 1007/ s00401‑ 018‑ 1855‑6 autonomic higher order processing nuclei of the lower brain stem are 29. Kaufman SK, Sanders DW, Thomas TL, Ruchinskas AJ, Vaquer‑Alicea among the early targets of the Alzheimer’s disease‑related cytoskeletal J, Sharma AM, Miller TM, Diamond MI (2016) Tau prion strains dictate pathology. Acta Neuropathol 101:555–564. https:// doi. org/ 10. 1007/ s0040 patterns of cell pathology, progression rate, and regional vulnerability 10000 320 In Vivo. Neuron 92:796–812. https:// doi. org/ 10. 1016/j. neuron. 2016. 09. 055 47. Sanders DW, Kaufman SK, DeVos SL, Sharma AM, Mirbaha H, Li A, Barker 30. Kaufman SK, Thomas TL, Del Tredici K, Braak H, Diamond MI (2017) Char‑ SJ, Foley AC, Thorpe JR, Serpell LC et al (2014) Distinct tau prion strains acterization of tau prion seeding activity and strains from formaldehyde‑ propagate in cells and mice and define different tauopathies. Neuron fixed tissue. Acta Neuropathol Commun 5:41. https:// doi. org/ 10. 1186/ 82:1271–1288. https:// doi. org/ 10. 1016/j. neuron. 2014. 04. 047 s40478‑ 017‑ 0442‑8 48. Sanders DW, Kaufman SK, Holmes BB, Diamond MI (2016) Prions and 31. Kovacs GG (2020) Astroglia and tau: new perspectives. Front Aging Neu‑ Protein Assemblies that Convey Biological Information in Health and rosci 12:96. https:// doi. org/ 10. 3389/ fnagi. 2020. 00096 Disease. Neuron 89:433–448. https:// doi. org/ 10. 1016/j. neuron. 2016. 01. 32. Kovacs GG, Ferrer I, Grinberg LT, Alafuzoff I, Attems J, Budka H, Cairns NJ, 026 Crary JF, Duyckaerts C, Ghetti B et al (2016) Aging‑related tau astro ‑ 49. Schön C, Hoffmann NA, Ochs SM, Burgold S, Filser S, Steinbach S, Seeliger gliopathy (ARTAG): harmonized evaluation strategy. Acta Neuropathol MW, Arzberger T, Goedert M, Kretzschmar HA et al (2012) Long‑term 131:87–102. https:// doi. org/ 10. 1007/ s00401‑ 015‑ 1509‑x in vivo imaging of fibrillar tau in the retina of P301S transgenic mice. PLoS 33. Kovacs GG, Lukic MJ, Irwin DJ, Arzberger T, Respondek G, Lee EB, Cough‑ ONE 7:e53547. https:// doi. org/ 10. 1371/ journ al. pone. 00535 47 lin D, Giese A, Grossman M, Kurz C et al (2020) Distribution patterns 50. Singleton AB, Wharton A, O’Brien KK, Walker MP, McKeith IG, Ballard of tau pathology in progressive supranuclear palsy. Acta Neuropathol CG, O’Brien J, Perry RH, Ince PG, Edwardson JA et al (2002) Clinical and 140:99–119. https:// doi. org/ 10. 1007/ s00401‑ 020‑ 02158‑2 neuropathological correlates of apolipoprotein E genotype in dementia 34. Kovacs GG, Milenkovic I, Wohrer A, Hoftberger R, Gelpi E, Haberler C, with Lewy bodies. Dement Geriatr Cogn Disord 14:167–175. https:// doi. Honigschnabl S, Reiner‑ Concin A, Heinzl H, Jungwirth S et al (2013) org/ 10. 1159/ 00006 6022 Non‑Alzheimer neurodegenerative pathologies and their combina‑ 51. Sun A, Nguyen XV, Bing G (2002) Comparative analysis of an improved tions are more frequent than commonly believed in the elderly brain: thioflavin‑s stain, Gallyas silver stain, and immunohistochemistry for a community‑based autopsy series. Acta Neuropathol 126:365–384. neurofibrillary tangle demonstration on the same sections. J Histochem https:// doi. org/ 10. 1007/ s00401‑ 013‑ 1157‑y Cytochem 50:463–472. https:// doi. org/ 10. 1177/ 00221 55402 05000 403 35. Lee VM, Goedert M, Trojanowski JQ (2001) Neurodegenerative tauopa‑ 52. Tan RH, Shepherd CE, Kril JJ, McCann H, McGeachie A, McGinley C, Affleck thies. Annu Rev Neurosci 24:1121–1159. https:// doi. org/ 10. 1146/ annur ev. A, Halliday GM (2013) Classification of FTLD ‑ TDP cases into pathological neuro. 24.1. 1121 subtypes using antibodies against phosphorylated and non‑phospho ‑ 36. Lowe VJ, Wiste HJ, Senjem ML, Weigand SD, Therneau TM, Boeve BF, rylated TDP43. Acta Neuropathol Commun 1:33. https:// doi. org/ 10. 1186/ Josephs KA, Fang P, Pandey MK, Murray ME et al (2018) Widespread brain 2051‑ 5960‑1‑ 33 tau and its association with ageing, braak stage and Alzheimer’s demen‑ 53. Tsuang D, Leverenz JB, Lopez OL, Hamilton RL, Bennett DA, Schneider tia. Brain 141:271–287. https:// doi. org/ 10. 1093/ brain/ awx320 JA, Buchman AS, Larson EB, Crane PK, Kaye JA et al (2013) APOE epsilon4 37. McKee AC, Cairns NJ, Dickson DW, Folkerth RD, Keene CD, Litvan I, Perl DP, increases risk for dementia in pure synucleinopathies. JAMA Neurol Stein TD, Vonsattel JP, Stewart W et al (2016) The first NINDS/NIBIB con‑ 70:223–228. https:// doi. org/ 10. 1001/ jaman eurol. 2013. 600 sensus meeting to define neuropathological criteria for the diagnosis of 54. Uchihara T (2007) Silver diagnosis in neuropathology: principles, practice chronic traumatic encephalopathy. Acta Neuropathol 131:75–86. https:// and revised interpretation. Acta Neuropathol 113:483–499. https:// doi. doi. org/ 10. 1007/ s00401‑ 015‑ 1515‑zorg/ 10. 1007/ s00401‑ 007‑ 0200‑2 38. Mercken M, Vandermeeren M, Lubke U, Six J, Boons J, Van de Voorde 55. Wennberg AM, Tosakulwong N, Lesnick TG, Murray ME, Whitwell JL, A, Martin JJ, Gheuens J (1992) Monoclonal antibodies with selective Liesinger AM, Petrucelli L, Boeve BF, Parisi JE, Knopman DS et al (2018) specificity for Alzheimer Tau are directed against phosphatase ‑sensitive Association of Apolipoprotein E epsilon4 With Transactive Response epitopes. Acta Neuropathol 84:265–272. https:// doi. org/ 10. 1007/ BF002 DNA‑Binding Protein 43. JAMA Neurol 75:1347–1354. https:// doi. org/ 10. 278191001/ jaman eurol. 2018. 3139 39. Mirbaha H, Chen D, Morazova OA, Ruff KM, Sharma AM, Liu X, Goodarzi 56. Xia C, Makaretz SJ, Caso C, McGinnis S, Gomperts SN, Sepulcre J, M, Pappu RV, Colby DW, Mirzaei H et al (2018) Inert and seed‑ competent Gomez‑Isla T, Hyman BT, Schultz A, Vasdev N et al (2017) Association tau monomers suggest structural origins of aggregation. Elife. https:// doi. of in vivo [18F]AV‑1451 tau PET imaging results with cortical atrophy org/ 10. 7554/ eLife. 36584 S topschinski et al. acta neuropathol commun (2021) 9:164 Page 19 of 19 and symptoms in typical and atypical Alzheimer disease. JAMA Neurol Publisher’s Note 74:427–436. https:// doi. org/ 10. 1001/ jaman eurol. 2016. 5755 Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 57. Yoshiyama Y, Higuchi M, Zhang B, Huang SM, Iwata N, Saido TC, Maeda lished maps and institutional affiliations. J, Suhara T, Trojanowski JQ, Lee VM (2007) Synapse loss and microglial activation precede tangles in a P301S tauopathy mouse model. Neuron 53:337–351. https:// doi. org/ 10. 1016/j. neuron. 2007. 01. 010 Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your ﬁeld rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals http://www.deepdyve.com/lp/springer-journals/anatomic-survey-of-seeding-in-alzheimer-s-disease-brains-reveals-eNZOeSAkpm

Loading next page...

References (58)

(2020)
Distribution patterns of tau pathology in progressive supranuclear palsy
Acta Neuropathol, 140
(2016)
Tau prion strains dictate patterns of cell pathology, progression rate, and regional vulnerability In Vivo
Neuron, 92
(2018)
Association of Apolipoprotein E epsilon4 With Transactive Response DNA-Binding Protein 43
JAMA Neurol, 75
(2015)
PART is part of Alzheimer disease
Acta Neuropathol, 129
(2018)
Aging-related tau astrogliopathy (ARTAG): not only tau phosphorylation in astrocytes
Brain Pathol, 28
(2016)
Tau PET patterns mirror clinical and neuroanatomical variability in Alzheimer's disease
Brain, 139
(2015)
Neuroanatomy and pathology of sporadic Alzheimer's disease
Adv Anat Embryol Cell Biol, 215
(1998)
Improved method facilitates reliable APOE genotyping of genomic DNA extracted from formaldehyde-fixed pathology specimens
J Neurosci Methods, 79
(2015)
Early stage of progressive supranuclear palsy: a neuropathological study of 324 consecutive autopsy cases
J Nippon Med Sch, 82
(2015)
Tau trimers are the minimal propagation unit spontaneously internalized to seed intracellular aggregation
J Biol Chem, 290
(2012)
Long-term in vivo imaging of fibrillar tau in the retina of P301S transgenic mice
PLoS ONE, 7
(2005)
Olfactory involvement in aging and Alzheimer's disease: an autopsy study
J Alzheimers Dis, 7
(1991)
Neuropathological stageing of Alzheimer-related changes
Acta Neuropathol, 82
(2003)
Staging of brain pathology related to sporadic Parkinson's disease
Neurobiol Aging, 24
(2013)
Classification of FTLD-TDP cases into pathological subtypes using antibodies against phosphorylated and non-phosphorylated TDP43
Acta Neuropathol Commun, 1
(2018)
Tau seeding activity begins in the transentorhinal/entorhinal regions and anticipates phospho-tau pathology in Alzheimer's disease and PART
Acta Neuropathol, 136
(2002)
Progressive supranuclear palsy: neuronal and glial cytoskeletal pathology in the higher order processing autonomic nuclei of the lower brainstem
Neuropathol Appl Neurobiol, 28
(2014)
Proteopathic tau seeding predicts tauopathy in vivo
Proc Natl Acad Sci U S A, 111
(2012)
National institute on Aging-Alzheimer's association guidelines for the neuropathologic assessment of Alzheimer's disease
Alzheimers Dement, 8
(2015)
Longitudinal assessment of tau pathology in patients with Alzheimer's disease using [18F]THK-5117 positron emission tomography
PLoS ONE, 10
(2001)
Neurodegenerative tauopathies
Annu Rev Neurosci, 24
(2020)
Astroglia and tau: new perspectives
Front Aging Neurosci, 12
(2017)
Association of in vivo [18F]AV-1451 tau PET imaging results with cortical atrophy and symptoms in typical and atypical Alzheimer disease
JAMA Neurol, 74
(2016)
Aging-related tau astrogliopathy (ARTAG): harmonized evaluation strategy
Acta Neuropathol, 131
(2013)
Non-Alzheimer neurodegenerative pathologies and their combinations are more frequent than commonly believed in the elderly brain: a community-based autopsy series
Acta Neuropathol, 126
(1998)
Neuropil threads are collinear with MAP2 immunostaining in neuronal dendrites of Alzheimer brain
J Neuropathol Exp Neurol, 57
(2018)
Networks of tau distribution in Alzheimer's disease
Brain, 141
(1989)
Cloning and sequencing of the cDNA encoding an isoform of microtubule-associated protein tau containing four tandem repeats: differential expression of tau protein mRNAs in human brain
EMBO J, 8
(2021)
Ultrasensitive tau biosensor cells detect no seeding in Alzheimer's disease CSF
Acta Neuropathol Commun, 9
(2007)
Silver diagnosis in neuropathology: principles, practice and revised interpretation
Acta Neuropathol, 113
(2018)
Widespread brain tau and its association with ageing, braak stage and Alzheimer's dementia
Brain, 141
(2014)
Distinct tau prion strains propagate in cells and mice and define different tauopathies
Neuron, 82
(2002)
Comparative analysis of an improved thioflavin-s stain, Gallyas silver stain, and immunohistochemistry for neurofibrillary tangle demonstration on the same sections
J Histochem Cytochem, 50
(2007)
Synapse loss and microglial activation precede tangles in a P301S tauopathy mouse model
Neuron, 53
(2014)
Primary age-related tauopathy (PART): a common pathology associated with human aging
Acta Neuropathol, 128
(2017)
Widespread tau seeding activity at early Braak stages
Acta Neuropathol, 133
(2015)
Synaptic contacts enhance cell-to-cell tau pathology propagation
Cell Rep, 11
R Nieuwenhuys (2008)
10.1007/978-3-540-34686-9
(2014)
Genetic analysis implicates APOE, SNCA and suggests lysosomal dysfunction in the etiology of dementia with Lewy bodies
Hum Mol Genet, 23
(2018)
Synaptic tau seeding precedes tau pathology in human Alzheimer's disease brain
Front Neurosci, 12
(2002)
Clinical and neuropathological correlates of apolipoprotein E genotype in dementia with Lewy bodies
Dement Geriatr Cogn Disord, 14
(2019)
How to demix Alzheimer-type and PSP-type tau lesions out of their mixture -hybrid approach to dissect comorbidity
Acta Neuropathol Commun, 7
(1992)
Monoclonal antibodies with selective specificity for Alzheimer Tau are directed against phosphatase-sensitive epitopes
Acta Neuropathol, 84
(2014)
Are cases with tau pathology occurring in the absence of Abeta deposits part of the AD-related pathological process?
Acta Neuropathol, 128
(2013)
APOE epsilon4 increases risk for dementia in pure synucleinopathies
JAMA Neurol, 70
(2016)
Prions and Protein Assemblies that Convey Biological Information in Health and Disease
Neuron, 89
(2018)
Inert and seed-competent tau monomers suggest structural origins of aggregation
Elife
(1993)
Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families
Science, 261
(2016)
Altered proteins in the aging brain
J Neuropathol Exp Neurol, 75
10.3791/53205
(1986)
Occurrence of neuropil threads in the senile human brain and in Alzheimer's disease: a third location of paired helical filaments outside of neurofibrillary tangles and neuritic plaques
Neurosci Lett, 65
(2017)
Characterization of tau prion seeding activity and strains from formaldehyde-fixed tissue
Acta Neuropathol Commun, 5
(2001)
The autonomic higher order processing nuclei of the lower brain stem are among the early targets of the Alzheimer's disease-related cytoskeletal pathology
Acta Neuropathol, 101
(Furman JL, Holmes BB, Diamond MI (2015) Sensitive Detection of Proteopathic Seeding Activity with FRET Flow Cytometry. J Vis Exp: e53205 10.3791/53205)
Furman JL, Holmes BB, Diamond MI (2015) Sensitive Detection of Proteopathic Seeding Activity with FRET Flow Cytometry. J Vis Exp: e53205 10.3791/53205
Furman JL, Holmes BB, Diamond MI (2015) Sensitive Detection of Proteopathic Seeding Activity with FRET Flow Cytometry. J Vis Exp: e53205 10.3791/53205, Furman JL, Holmes BB, Diamond MI (2015) Sensitive Detection of Proteopathic Seeding Activity with FRET Flow Cytometry. J Vis Exp: e53205 10.3791/53205
(2016)
The first NINDS/NIBIB consensus meeting to define neuropathological criteria for the diagnosis of chronic traumatic encephalopathy
Acta Neuropathol, 131
(2012)
Correlation of Alzheimer disease neuropathologic changes with cognitive status: a review of the literature
J Neuropathol Exp Neurol, 71
(2006)
Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry
Acta Neuropathol, 112
(2010)
Induction and spreading of tau pathology in a mouse model of Alzheimer's disease
Med Sci (Paris), 26

Publisher: Springer Journals
Copyright: Copyright © The Author(s) 2021
eISSN: 2051-5960
DOI: 10.1186/s40478-021-01255-x
Publisher site: See Article on Publisher Site

Abstract

Tauopathies are heterogeneous neurodegenerative diseases defined by progressive brain accumulation of tau aggre ‑ gates. The most common tauopathy, sporadic Alzheimer’s disease (AD), involves progressive tau deposition that can be divided into specific stages of neurofibrillary tangle pathology. This classification is consistent with experimental data which suggests that network‑based propagation is mediated by cell–cell transfer of tau “seeds”, or assemblies, that serve as templates for their own replication. Until now, seeding assays of AD brain have largely been limited to areas previously defined by NFT pathology. We now expand this work to additional regions. We selected 20 individu‑ als with AD pathology of NFT stages I, III, and V. We stained and classified 25 brain regions in each using the anti‑ phospho‑tau monoclonal antibody AT8. We measured tau seeding in each of the 500 samples using a cell‑based tau “biosensor” assay in which induction of intracellular tau aggregation is mediated by exogenous tau assemblies. We observed a progressive increase in tau seeding according to NFT stage. Seeding frequently preceded NFT pathology, e.g., in the basolateral subnucleus of the amygdala and the substantia nigra, pars compacta. We observed seed‑ ing in brain regions not previously known to develop tau pathology, e.g., the globus pallidus and internal capsule, where AT8 staining revealed mainly axonal accumulation of tau. AT8 staining in brain regions identified because of tau seeding also revealed pathology in a previously undescribed cell type: Bergmann glia of the cerebellar cortex. We also detected tau seeding in brain regions not previously examined, e.g., the intermediate reticular zone, dorsal raphe nucleus, amygdala, basal nucleus of Meynert, and olfactory bulb. In conclusion, tau histopathology and seeding are complementary analytical tools. Tau seeding assays reveal pathology in the absence of AT8 signal in some instances, and previously unrecognized sites of tau deposition. The variation in sites of seeding between individuals could underlie differences in the clinical presentation and course of AD. Keywords: Alzheimer’s disease, AT8, FRET biosensor, Neurofibrillary tangles, Prion propagation, Tau seeding, NFT staging Introduction Tauopathies are a heterogeneous group of neurodegen- erative diseases defined by progressive brain accumula - tion of tau aggregates [35]. Sporadic Alzheimer’s disease (AD) is the most common, and is uniquely defined by *Correspondence: Marc.diamond@UTSouthwestern.edu Barbara E. Stopschinski and Kelly Del Tredici have equally contributed to coexistent tau and amyloid β pathology. AD neuropa- this work. thology includes intraneuronal somatic and axonal pre- Center for Alzheimer’s and Neurodegenerative Diseases, Peter O’Donnell tangles and neurofibrillary tangles (NFTs), neuropil Jr. Brain Institute, NL10.120, University of Texas Southwestern Medical Center, 6000 Harry Hines Blvd., Dallas, TX 75390, USA threads (NTs), extraneuronal ghost tangles, and amyloid Full list of author information is available at the end of the article β plaques. Tau pathology progresses in a defined and Talitha Louise Thomas, BS (July 7, 1982 – October 28, 2020) © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 2 of 19 characteristic pattern, allowing AD classification into or insoluble tau protein could be observed [25]. Further, different stages that correlate with antemortem clinical in fresh frozen brain tissue from individuals with AD, presentation [4]. the assay also detected seeding prior to neuropathologi- Aggregated tau protein is often phosphorylated [35], cal changes [14, 25]. We subsequently refined the method and the anti-phospho-tau monoclonal antibody AT8 to quantify seeding in fixed brain sections, which was [26] is typically used for detection and staging. AT8 equally reliable [30]. In fixed tissues from multiple AD binds phospho-serine 202 and phospho-threonine 205 patients at different NFT stages, we observed tau seed - on aggregated tau protein, and marks AD intraneu- ing first in the TRE and EC rather than in the locus coer - ronal pathology (pretangles and NFTs) [38]. AT8 signal uleus (LC), as we had previously hypothesized based on increases with disease progression and allows the defi - AT8 histopathology [28]. All prior analyses have been nition of NFT stages [3, 6]. In NFT stage I, AT8 marks confined to regions known to contain NFT pathology in selected brainstem nuclei and the transentorhinal cortex AD. The extent of tau seeding across widespread brain (TRE). In NFT stage II, AT8 marks the entorhinal cor- regions, however, is unknown. In this work we have used tex (EC) in the parahippocampal gyrus. In NFT stage an optimized biosensor cell line (TauRD(P301S)v2H) III, AT8 marks the CA1 region of the hippocampus, and [23] and we have tested for seeding across a large cross- neocortical regions of the temporal neocortex adjacent section of brain regions with and without known NFT to the TRE. In NFT stages IV and V, AT8 marks neo- pathology. We have generated a map of AD brain across cortical regions including the superior temporal gyrus NFT stages. This has revealed surprising patterns, and a (STG), and in NFT stage VI it marks primary neocorti- new type of cellular tau pathology. cal areas such as the visual cortex (VC). At NFT stages III to IV, more than 50% of individuals have signs of mild cognitive impairment, whereas at NFT stages V and VI Methods more than 90% of individuals exhibit signs of moderate Generation of biosensor cell line (TauRD(P301S)v2H) to severe dementia [26]. The severity of AD dementia A second generation of high sensitivity biosensor cells correlates with the extent of postmortem tau pathology termed v2H has recently been produced [23]. Using the [41, 56]. Additionally, longitudinal tau PET imaging has previously described lentiviral FM5-YFP plasmid [47], we confirmed the progression of tau pathology along NFT inserted the tau segment 246 to 378 with the P301S stages, and its correlation with neuronal dysfunction and mutation, replaced the human ubiquitin C (Ubc) pro- neurodegeneration [24, 27, 36, 44]. moter with a human cytomegalovirus (CMV) promoter, Progressive tau aggregation in AD occurs in patterns and replaced the YFP sequence with an mCerulean3 or consistent with neural networks [4]. Recent data from mClover3 coding sequence. To reduce translation read in vitro and in vivo experimental systems is consistent through of the tau ATG start site and increase transgene with trans-neuronal spread of pathology similar to prion expression, the sequence upstream of tau was modified disease, in which pathological species move from cell to encode an optimal Kozak sequence (5’-GCC ACC to cell, serving as templates to convert native tau into a ACC ATG GCC-3’). The GCC after the ATG start codon pathogenic aggregation-prone form, and thereby propa- encodes the amino acid A246 in tau. The sequence link - gating tau pathology among connected brain regions ing the tau segment and the coding sequence of the fluo - [10, 11, 47, 48]. It is unknown whether this mechanism rophore (mCerulean3 or mClover3) was optimized to the underlies progression in humans, however the presence following sequence: 5’- GSAGSAAGSGEF-3’. of soluble, non-aggregated pretangle pathogenic tau To create the v2H line, low passage HEK293T cells (P5) “seeds” in human brain that anticipate the development were thawed and passaged with antibiotic free media of NFT pathology is very consistent with this idea [20, twice before co-administration of P301S 246–378 tau- 28]. mCerulean3 tau-mClover3 lentivirus. After four pas- To detect tau seeding in biological samples, we previ- sages, single cells were isolated via fluorescence activated ously developed a sensitive and specific cell-based “bio - cell sorting (FACS) based on low, intermediate, and high sensor” seeding assay, in which the tau repeat domain brightness levels for both mCerulean3 and mClover3. containing a single disease-associated mutation (P301S) Monoclonal colonies were cultured to high cell number is fused to complementary fluorescent proteins (e.g., and tested by seeding assays with recombinant fibrils and cyan/yellow; cerulean/clover; clover/ruby), and expressed AD lysate. The v2H line was chosen for low background in cells of choice. The fusion proteins aggregate upon signal and high sensitivity, and used in subsequent seed- exposure to tau seeds, which is quantified by flow cytom - ing experiments as a next-generation biosensor based on etry [19, 25]. In a transgenic mouse model, the seeding previously established protocols [25]. assay scored positive many months before histopathology S topschinski et al. acta neuropathol commun (2021) 9:164 Page 3 of 19 Culture of biosensor cells Human autopsy samples Stable monoclonal v2H FRET biosensor cells were Human autopsy tissue used for this study was obtained grown in complete media: Dulbecco’s Modified Eagle’s from n = 20 individuals (10 females, 10 males, age range Medium (DMEM) (Gibco) with 10% fetal bovine serum 50–93 years, Table 1) and 1 control (1 female, 30 years (Sigma), 1% penicillin/streptomycin (Gibco) and 1% of age) in compliance with ethics committee guidelines Glutamax (Gibco). Cells were cultured and passaged at the University of Ulm as well as German federal and at 37 °C, 5% CO , in a humidified incubator. Dulbecco’s state law governing human tissue usage. Informed writ- phosphate buffered saline (Life Technologies) was used ten consent for autopsy was obtained previously from for washing the cells prior to harvesting with 0.05% the patients or their next of kin. Brains were fixed Trypsin–EDTA (Life Technologies). in a 4% buffered aqueous solution of formaldehyde for 14 days. Tissue blocks from 25 brain regions were excised and embedded in polyethylene glycol (PEG Mouse breeding for positive and negative controls 1000, Merck, Carl Roth Ltd, Karlsruhe, Germany). All experiments involving animals were approved 100 μm serial sections were collected and stained free- by the University of Texas Southwestern Medical floating, as described previously [3 ] (Table 2). Brain tis- Center Institutional Animal Care and Use Committee sue and the remaining tissue sections were stored for (IACUC). All mice were housed under a 12 h light/dark subsequent use in a 4% aqueous solution of formalde- cycle, and were provided food and water ad libitum. We hyde at 8–15 °C for up to 26 years. used tau KO mice containing a GFP-encoding cDNA integrated into exon 1 of the MAPT gene as a negative control. These mice were obtained from Jackson Labo - APOE genotyping ratory and maintained on a C57BL/6 J background. Apolipoprotein E status was available for 16/20 of the As a positive control, we obtained transgenic mice individuals studied (Table 1). The ε4 allele is a major expressing 1N4R P301S human tau under the murine genetic risk factor for sporadic AD [12], TDP-43 pro- prion promoter [57] from Jackson Laboratory, and teinopathy [55] and for dementia with Lewy bodies (DLB) maintained them on a B6/C3 background. The posi - and Parkinson’s disease dementia [9, 50, 53]. APOE geno- tive control mice were anesthetized at age 2.5 months typing was performed (E.G.) using a semi-nested poly- with isoflurane and kept at 37 °C throughout the inocu - merase chain reaction assay and restriction isotyping lation. We used 10 μL gas-tight Hamilton syringes to with restriction enzyme HhaI [21]. Genomic DNA was inject 10 µg of clone 9 cell protein lysate (previously extracted from formaldehyde-fixed and paraffin-embed - described in [29, 47]) in the left hippocampus (bregma: ded brain specimens using the manufacturer’s protocols -2.5 mm posterior, -2 mm lateral, -1.8 mm ventral). The (QIAamp DNA Mini Kit, Qiagen, Hilden, Germany). mice were euthanized 4 weeks later as described below for seeding experiments. Neuropathological staging Neuropathological staging and disease classification of Mouse sample collection and preparation AD-associated pathology were performed (H.B., K.D.T.) The mice were anesthetized with isoflurane and per - according to a previously published modified Gallyas fused with chilled PBS + 0.03% heparin. Brains were silver-iodide staining protocol [3, 4] for recognition of post-fixed in 4% PFA overnight at 4 °C and placed in phosphorylated somatic argyrophilic (fibrillary) neuropil 30% sucrose in PBS until further use. Brains were sec- threads (NTs) [1, 5] and neurofibrillary tangles (NFTs), as tioned at 50 μm with a freezing microtome and placed well as of extraneuronal ghost tangles (‘tombstone’ tan- into cryoprotectant (32% ethylene glycol, 16% w/v gles) that display weak staining with the Gallyas method sucrose, in 50 mM phosphate buffer pH 7.4, stored at and strong staining with the Campbell-Switzer silver-pyr- -20 °C). 1 mm punches were then isolated from the left idine method. In addition, AT8 immunohistochemistry hippocampus using Miltex disposable punch biopsy (IHC, monoclonal anti-PHF-Tau antibody, 1:2000; Clone tools. Four 4 mm punches were placed into 100 µl AT8; Pierce Biotechnology [Thermo Scientific] Waltham, EDTA buffer (1 mM EDTA, pH 8.0), heated for 25 min MA [38], was performed. In contrast to the Gallyas at 95 °C and allowed to cool down for ~ 15 min. The method, AT8 IHC visualizes the broadest spectrum of samples were then sonicated with a water bath sonica- intraneuronal pathological tau: argyrophilic NFTs of the tor (Qsonica Q700MPX with chiller and tubing set) at Alzheimer type, NTs in dendritic processes, and non- 4 °C at 50 amplitude for 60 min, and stored at -80 °C argyrophilic axonal aggregates and pretangles. AT8 IHC until further use. Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 4 of 19 Table 1 Demographic and neuropathological data Case f/m age brain wt NFT Aβ α-syn TDP-43 APOE Diagnoses 1 f 55 1455 I 0 0 − ε3/4 intraventricular hemorrhage 2 m 55 1780 I 0 0 − ε3/3 bronchopneumonia 3 f 50 1293 I 0 0 − ε3/3 PCOM aneurysm, SAB 4 m 50 1600 I 0 0 − ε4/4 M. Werlhof, hepatitis C 5 f 72 1070 I 3 0 − ε3/4 cardiac failure 6 f 79 1165 I 0 0 − ε3/3 coronary artery disease 7 f 93 1090 III (IV ) 2 0 − na craniocerebral trauma 8 m 68 1380 III 0 0 − na aspiration pneumonia 9 m 72 1500 III 0 0 − ε3/4 cardiac failure 10 m 74 1465 III 0 0 − ε4/4 malignant neoplasm 11 m 76 1520 III 4 0 + na pneumonia 12 f 81 1130 III 2 0 − ε3/3 acute myeloid leukemia 13 f 88 1495 III 3 0 − ε3/4 cerebral hemorrhage parietooccipital 14 m 84 1170 V ( VI) 3 0 + ε3/4 AD 15 f 84 1175 V 3 2 + ε3/3 AD, ILBD 16 m 58 1335 V 2 0 − ε3/4 AD, myocardial infarction 17 f 72 1185 V 4 0 − ε3/3 AD, craniocerebral trauma after falling 18 f 76 1405 V 4 0 − ε3/4 AD, craniocerebral trauma after falling 19 m 76 1205 V 5 0 − na AD, cardiac failure 20 m 78 1460 V 5 0 − ε3/4 AD Control 21 f 30 1315 0 0 0 − ε3/3 malignant neoplasm 20 cases fall into three NFT groups: NFT stage I (4 females, 2 males, 50–79 years); NFT stage III (3 females, 4 males, 68–93 years); NFT stage V (3 females, 4 males, 58–84 years). Abbreviations: f, m—female, male; age—age in years; brain wt—fresh brain weight in grams; NFT—Alzheimer’s disease-related neurofibrillary tangle stage using Gallyas silver-iodide staining; Aβ—amyloid-β deposition phase using 4G8 IHC; α-syn—Parkinson disease-related neuropathological stage using α-synuclein IHC; TDP-43—43-kDa TAR DNA-binding protein neuronal inclusions; APOE—APOE allele status; n/a -not available; AD—Alzheimer’s disease; ILBD— incidental Lewy body disease; PCOM -posterior communicating artery; SAB—subarachnoid bleeding detects ghost tangles less effectively than Gallyas silver polyclonal rabbit antibody recognizing the N-terminal of staining, or not at all. The character and relative merits of normal TDP-43 (1:5000; Proteintech, Manchester, UK) thioflavin-S staining, Gallyas and Campbell-Switzer silver [52]. We staged all cases for sporadic Parkinson’s dis- staining, as well as more conventional silver methods (the ease (PD), as described elsewhere [8] (Table 1). One case modified Bielschowsky and the Bodian methods) in rela - showed incidental α-synuclein-positive Lewy pathol- tion to tau isoforms and to IHC have been discussed in ogy; three cases displayed coincident TDP-43 immu- detail elsewhere [51, 54]. We evaluated the presence of noreactivity [17, 34] (Table 1). Histological slides were aging-related tau astrogliopathy (ARTAG) as proposed viewed with an Olympus BX61 microscope (Olympus by Kovacs [32]. We visualized and staged Aβ deposition Optical, Tokyo, Japan). Pathology was assessed semi- using the monoclonal anti-Aβ antibody 4G8 (1:5000; quantitatively on a four-point scale: 0 = no detect- Clone 4G8; BioLegend, San Diego, CA) as recommended able tau inclusions, + = mild (at least one or two previously [26]. Clinical AD classification included cases AT8-positive cell soma/somata); + + = moderate inclu- with tau stages III/V and Aβ phases ≥ 2 [13, 15] (Table 1). sions; + + + = severe inclusions. Digital micrographs We excluded other non-AD tauopathies, including of IHC-stained sections (Figs. 4, 5) were taken with an argyrophilic grain disease, progressive supranuclear Olympus XC50 camera (H.B.) using the Cell D Imaging palsy, Pick’s disease, corticobasal degeneration, and Software (Olympus, Münster, Germany). The extended Niemann-Pick disease type C. Separate sets of 100 µm focal imaging (EFI) function was used for stacking free-floating sections from all cases were immunostained images at different optical planes (Cell D Imaging Soft - using the following primary antibodies: (1) a monoclo- ware, Olympus, Münster, Germany). The EFI algorithm nal anti-syn-1 antibody (1:2000; Clone number 42; BD extracts the image features with the sharpest contrast Biosciences, Eysins, Switzerland) for detection of Lewy from all layers of the stack and merges them into a single body disease-related α-synuclein inclusions [26]; (2) a image. S topschinski et al. acta neuropathol commun (2021) 9:164 Page 5 of 19 Table 2 Brain regions sampled punch tool from positive and negative control mice (S.E.). Regions 1 Transentorhinal cortex ( TRE) Human sample preparation 2 Entorhinal cortex (EC, Brodmann Area 28) One 4 mm punch or two 3 mm punches were placed 3 Ammon’s horn, sector 1 (CA1, hippocampal formation) into 100 µl EDTA buffer (1 mM EDTA, pH 8.0), heated 4 Amygdala, basolateral subnucleus (AMY ) for 25 min at 95 °C and allowed to cool down at 4 °C 5 Superior (first) temporal gyrus (STG, Brodmann Area 22) for ~ 15 min. The samples were then sonicated with a 6 Transverse temporal gyrus of Heschl ( TTG, Brodmann Area 41) water bath sonicator (Qsonica Q700MPX with chiller 7 Primary visual neocortex (PV, Brodmann Area 17) and tubing set) at 4 °C at 50 amplitude for 60 min, and 8 Peristriate neocortex, high order sensory neocortex (Brodmann stored at −80 °C until further use. Area 19) 9 Anterior cingulate cortex—skeletomotor/emotion‑autonomic integration (ACC, Brodmann Areas 24/32) Transduction of biosensor cell lines, flow cytometry 10 Retrosplenial/posterior cingulate cortex—memory/visuospatial and seeding analysis orientation (RSC/PCC, Brodmann Areas 23/29/30) The seeding assay was conducted as previously 11 Putamen (PUT ) described with the following changes: biosensor cells 12 Globus pallidus (GP) were plated at a density of 25,000 cells/well in a 96-well 13 Mediodorsal complex of thalamus (MD) plate in a media volume of 130 µl per well. The mouse 14 Orbitofrontal cortex (OFC, Brodmann Area 11) and human tissue samples were thawed on ice, fol- 15 Substantia nigra, pars compacta (SNpc) lowed by thorough vortexing and incubation with Lipo- 16 Locus coeruleus (LC) fectamine 2000 for 30 min. 1 µl of tissue lysate with 17 Basal nucleus of Meynert(BN) 0.5 µl of lipofectamine and 18.5 µl of OptiMEM (Gibco, 18 Pontine gray (PG) Life Technologies) was added to each well, resulting 19 Inferior olivary nucleus (IO) in 20 µl total. For each experiment, negative controls 20 Cerebellar cortex (CC) received either Lipofectamine in OptiMEM (lipo- 21 Cerebellar dentate nucleus (DN) fectamine controls), or OptiMEM (buffer controls). 22 Internal capsule, anterior limb (IC) The lysate-lipofectamine mix was applied to the cells, 23 Terminal stria ( TS) and cells were incubated for an additional 72 h. Cells 24 Olfactory bulb (OB) were harvested with 0.05% trypsin and fixed in 2% PFA 25 Optic chiasm/tract (OC) for 10 min, then resuspended in flow cytometry buffer Punch biopsies were made from unstained sections of the 25 brain regions (HBSS plus 1% FBS and 1 mM EDTA). An LSRFortessa shown above using a 4 mm (3 mm for regions 19 and 22 in Experiment II) punch biopsy tool. Cross-contamination of seeding activity between individuals and SORP (BD Biosciences) was used to perform FRET regions was prevented by disposing the biopsy tool after each punch flow cytometry. We quantified FRET as previously described with the following modification: we identi - fied single cells that were double-positive for mCeru - Punch samples lean and mClover and subsequently quantified FRET From each case, including the negative human control, positive cells within this population. For each data punch samples were collected (K.D.T.) free-floating set, 3 technical replicates were included. Data analy- from unstained sections of the 25 brain regions shown sis was performed using FlowJo v10 software (Treestar in Table 2 with a punch biopsy tool (Kai Industries Co, Inc.), GraphPad Prism v8.4.3 for Mac OS X, and Excel Ltd. Japan) – with diameter of either 4 mm (result- 2 v16.16.25 (Microsoft). ing in a punch volume of ~ 1.257 mm ) or 3 mm (with estimated punch volume of ~ 0.706 mm ). The 3 mm punch device was only used for the internal capsule Statistical analyses (IC) and the inferior olivary nucleus (IO) in the 2nd Samples were collected at the University of Ulm and set of punches to ensure that the punches were con- cases were blinded prior to seeding analyses by B.E.S. at fined to the immediate target regions. To avoid cross UT Southwestern Medical Center. Flow cytometry gat- contamination of seeding between individuals and ing and analysis of seeding were completed prior to the regions, punch tools were used only once. Samples decoding and interpretation of the seeding results. All were encoded and all subsequent preparation and seed- statistical analysis was performed using GraphPad Prism ing assays were performed in a blinded fashion. Tissue v8.4.3 for Mac OS X and Excel v16.16.25 (Microsoft). punches were stored in 1 × TBS at 4 °C until use. Brain Statistical significance between seeding at different NFT tissue was collected in the same way with the 4 mm Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 6 of 19 stages was determined by performing a non-parametric the seeding assay was appropriately detecting tau seed- rank order test (Whitney-Mann test). Correlation analy- ing only when present and not detecting seeding when sis (linear regression) was performed and Spearman r tau was absent in human tissue. Therefore, we did not correlation was calculated to test the reproducibility of sample each of the 25 brain regions from case 21. We seeding between different experimental runs. used the average of all negative samples plus 3 standard deviations to determine the “positive” seeding thresh- old. We considered samples with seeding of 0.35% Generation of 3D tau seeding map and above to be positive, whereas samples below this Magnetic resonance imaging of the brain was performed threshold were considered negative. With a 3 SD cut- in a healthy volunteer using a 3 T Siemens Prisma off, we increased the specificity of our analysis at the MRI scanner, including acquisition of 3D T1-weighted cost of sensitivity. The coefficient of variation (defined MPRAGE sequence with 1 mm isotropic resolution. as standard deviation divided by the FRET average) was Segmentation of regions of interest (ROIs) within the used as a measure of precision and is high for samples brain was performed semi-automatically using Free- below the seeding threshold and low for samples above Surfer (version 5.3.0, http:// surfer. nmr. mgh. harva rd. edu). the threshold (Fig. 1). Given the relatively high speci- In cases where ROIs did not already exist in the Free- ficity of the assay, negative results do not rule out tau Surfer library, manual segmentation was performed by seeding in a given sample. an experienced board-certified neuroradiologist (F.F.Y.) using the Segmentation Editor tool in 3D Slicer (version 4.10.2, https:// www. slicer. org/). The weighted average tau Reproducibility between experimental runs and samples deposition from all subjects within each NFT stage group To test seeding within regions, we selected 50 punches was then applied to each ROI using a customized script from 20 individuals and obtained a second set of punches in Matlab (version R2015b). Brain regions were included from the same section as the 1st set whenever possible. only if at least one subject within each NFT stage exhib- If not possible (because of limited amount of tissue), the ited seeding. The Build Surface function within Mango 2nd punch was obtained from an immediately adjacent (version 4.1, http:// ric. uthsc sa. edu/ mango/) was then section. We then performed 2 additional seeding experi- used to visualize the ROIs within a 3D projection of a ments: In experiment II, the 2nd sample set was tested control brain. (run 1). In experiment III, both sample sets were thawed a second time and tested for seeding (run 2). Thus, the Results 1st sample set was tested after one freeze thaw cycle in Sampling of 25 brain regions across 20 individuals experiment I (run 1), and after a second freeze–thaw Previous publications from our group and others have cycle in experiment III (run 2). In the same way, the 2nd studied seeding in a limited number of brain samples sample set was tested in experiment II (run 1) and experi- from AD patients using the biosensor system [14, 20, 25, ment III (run 2) after one versus two freeze–thaw cycles 28]. We chose 20 individuals with confirmed AD pathol - (Fig. 2). For both experiments II and III, the coefficient ogy for this analysis (Table 1). Given that the differences of variation above the previously defined seeding thresh - in tau tangle pathology between NFT stage I/II, III/IV, old of 0.35% was low (Fig. 2b and c). We then used linear and V/VI are subtle, we limited our study to NFT stages regression to compare the sample sets and experimental I, III and V. Furthermore, we used the punch device runs. Seeding correlated well between different experi - established by Kaufman et al. 2018 [28] for more precise mental runs (run 1 and 2) of the same samples with R sampling of the 25 brain regions of interest as opposed to in the range of 0.7 – 0.8 (Fig. 2d, e). The reproducibility sampling by dissecting large tissue pieces (Table 2). of the seeding data between different samples (1st versus 2nd sample set) had low reliability (Fig. 2f, g). Seeding threshold determination To determine the lower limit of detection, tissue lysate was transduced into v2H biosensor cells. We quanti- Progressive accumulation of seeding within individuals fied the percentage of FRET positive cells on the flow To examine the progression of seeding across all brain cytometer as a correlate of intracellular tau seeding, regions, we created a heat map with the seeding for each compared to negative control samples. Negative sam- individual and each brain region (Fig. 3a). We also plotted ples included human tau-negative brain tissue from seeding for each individual brain region for all 20 indi- case 21 (taken from the pons), brain tissue from tau viduals (Fig. 3b) and generated a 3D seeding map for each knockout mice, and wells treated with lipofectamine NFT stage (Additional file 2: Figure S6). In general, seed- or buffer only. Note that the samples from case 21 ing increased with higher NFT stages in all brain regions were included as internal assay control to ensure that S topschinski et al. acta neuropathol commun (2021) 9:164 Page 7 of 19 Fig. 1 Seeding profile of cases (Experiment I). Punch biopsies were taken from 25 brain regions in 20 individuals (NFT stages I, III and V ), homogenized, and transduced into v2H biosensor cells. Seeding was quantified by determining the percentage of FRET positive cells on a flow cytometer. Each sample was tested in biological/technical triplicate, and the average is reported. a Negative controls included cells that were treated with: (1) lipofectamine (+ buffer); (2) buffer only; (3) tau negative human brain tissue (from individual number 21 in Table 1; 4) brain tissue from tau knockout mice. The average seeding for each condition is shown as percentage of FRET positive cells ± standard deviation. We used the average of all negative samples and 3 × their respective standard deviations to determine the seeding threshold at 0.35% (in red). Only samples with seeding above 0.35% were scored positive. b The FRET average for each sample in experiment 1 was plotted on a log scale against the coefficient of variation (measure of assay precision defined as the standard deviation divided by the FRET average). The coefficient of variation is low for samples with seeding above the threshold of ~ 0.35%. For samples with seeding below this threshold, the coefficient of variation is significantly larger. c The FRET average for controls with standard deviation was plotted on a linear scale against the coefficient of variation. For lipofectamine and buffer controls, the average and standard deviation were derived from all 97 respectively 12 wells in this experiment. For tau negative human tissue and tau knockout mouse tissue, averages and standard deviation were calculated for each of 2 triplicates and plotted separately. Note that all controls ± standard deviation are below the seeding threshold of 0.35%. Color code: lipofectamine (yellow), buffer (pink), tau negative human brain tissue (blue), tau knockout mouse brain tissue (green), samples from individuals 1–20 (black) examined. For unexpected seeding results (Table 3), we (n = 4/6) had positive seeding in the TRE. In 2 cases performed AT8 staining on selected brain regions to test (cases 1 and 3), seeding in the TRE was below the for tau deposition (Additional file 1: Table S1). threshold of detection despite the presence of 9 and 1 AT8-positive neuron(s) respectively in this region Tau seeding starts in the transentorhinal cortex (Additional file 1 : Table S1 and Fig. 4a). We detected seeding in the transentorhinal cor- tex (TRE) in all individuals at NFT stages III and V (Fig. 3a). At NFT stage I, the majority of individuals Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 8 of 19 ab Fig. 2 Reproducibility of seeding data. a To determine variation in seeding by region, we randomly selected a subset of 50 samples from the 1st sample set (tested in Exp I and shown in Fig. 1) and gathered a second set of punches from these brain regions (2nd sample set). Exp II tested seeding in the 2nd sample set (50 samples total plus controls). In Experiment III, seeding of both sample sets was tested a second time (100 samples total plus controls). In summary, both 1st and 2nd sample sets were tested in 2 separate runs, with the second run following a freeze–thaw cycle (Run 1 and 2). b and c show the plots for coefficient of variation versus FRET average for samples and controls in experiments II and III. The graphs were generated in the same way as for Experiment I in Fig. 1b and c. All control samples fall below the seeding threshold of 0.35%. Color code: lipofectamine (yellow), buffer (pink), tau negative human brain tissue (blue), tau knockout mouse brain tissue (green), samples from individuals 1–20 (black). d and e show a correlation analysis (linear regression) to test the reproducibility of seeding for the same sample sets between different experimental runs. f and g show the correlation analysis to test the reproducibility between different punches from the same brain region. Note that control samples were not included in graphs d to g Early seeding in the entorhinal cortex, CA1, amygdala, seeding. The entorhinal cortex (EC) and CA1/hippocam - and locus coeruleus pal formation scored positive in n = 2/6 NFT stage I Four additional brains regions demonstrated early cases, and remained positive at all higher NFT stages S topschinski et al. acta neuropathol commun (2021) 9:164 Page 9 of 19 Fig. 3 Seeding in 20 individuals, 25 brain regions (derived from Experiment I). a Seeding data heat map: Data points below the seeding threshold of 0.35% are colored in gray. Data points equal and above the seeding threshold are shaded with a graded color scale ranging from yellow (low) to red (high). Seeding data from human control brain (tau negative) was included as a comparison. b Seeding data from 25 brain regions plotted as individual graphs and separated according to NFT stage. Individual symbols (dot, square, triangle) represent data from individuals at each NFT stage. Statistical significance was determined by performing a non‑parametric rank order test ( Whitney‑Mann test) to compare NFT I vs. III, III vs. V, and I vs. V. ns = non‑significant, *p < 0.05, **p < 0.01. Errors bars show SD (n = 14/14) (Fig. 3a). Seeding was also detected in n = 1/6 at later stages (NFT V and VI) [4]. AT8 staining in n = 5 NFT stage I individuals in the basolateral subnucleus of seeding-positive cases (e.g., at NFT stages I and III, Addi- the amygdala (AMY), and this region remained positive tional file 1: Table S1) revealed immunopositive SNpc in the vast majority of NFT III (n = 6/7) and all NFT V axons in all individuals examined (Fig. 5e). However, cases (n = 7/7) (Figs. 3a and 5g). The locus coeruleus (LC) seeding above threshold could only be detected in one exhibited seeding in n = 1/6 at NFT stage I, and at NFT case (case 12) that also displayed AT8-positive neurons stages III and V the LC was positive for seeding in all (Additional file 1: Table S1). individuals tested, 2 of whom displayed strong seeding at The globus pallidus (GP) does not typically show AD- NFT stage V (Figs. 3a and 5h, i). This is broadly consist - associated tau pathology [7]. However, n = 4/7 cases at ent with our prior work [28]. NFT stage III and all cases at NFT stage V demonstrated seeding (Fig. 3). AT8 staining in 9 individuals revealed Intermediate seeding in 9 brain regions AT8 positivity in axons but not nerve cell somata (Addi- The superior temporal gyrus (STG), the peristriate neo - tional file 1: Table S1). In two separate punches from case cortex (area 19, PS), and the terminal stria (TS) scored 7 (Fig. 4g and h), we found a single AT8-positive neuronal positive for seeding in some individuals (n = 3/7) at NFT body in the GP. Seeding-positive axons in the GP could stage III, and in all individuals at NFT stage V (Fig. 3a). have their origins in the basal nucleus of Meynert (BN), The anterior cingulate cortex (ACC) demonstrated posi - which displayed seeding in n = 4/7 cases at NFT stage III tive seeding in most individuals at NFT stage III (n = 5/7) and in n = 7/7 NFT stage V cases (Fig. 3). and all NFT stage V cases, and the orbitofrontal cortex The olfactory bulb (OB) scored positive in n = 9/19 (OFC) as well as Meynert’s basal nucleus (BN) were seed- cases (Fig. 3a). We found no evidence of tau seeding at ing-positive in n = 4/6 cases at NFT stage III and in all NFT stage I (Fig. 3a). At NFT stage III, n = 4/7 individu- individuals at NFT stage V (Fig. 3a). als displayed seeding in this region, and at NFT stage V, We detected robust seeding in the substantia nigra, n = 5/6 individuals scored positive, 1 of which was par- pars compacta (SNpc) in most cases (n = 6/7) at NFT ticularly pronounced (case 15). No OB tissue from case stage III and n = 7/7 at NFT stage V (Fig. 3a). By contrast, 16 was available. tangle pathology is typically observed in this region only Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 10 of 19 Fig. 3 continued S topschinski et al. acta neuropathol commun (2021) 9:164 Page 11 of 19 Late seeding in 5 brain regions because of its close proximity to the intermediate reticu- Seeding at late NFT stages was present in 5 brain regions. lar zone (IRZ) (Figs. 4d, e and 5a-d), which can contain Heschl’s gyrus (TTG), the putamen (PUT), retrosplenial/ AT8-positive axons and/or neurons in early AD stages posterior cingulate cortex (RSC/PCC) and mediodorsal [46]. While we obtained mostly clean punches from IO, complex of thalamus (MD) were positive for seeding in we cannot exclude that a few contained edges of the only n = 1/7 to n = 2/7 individuals at NFT stage III, and IRZ, as noted in Additional file 1: Table S1 and shown in in most or all individuals at NFT stage V (Fig. 3). The pri - Fig. 5a and b. Use of a 3 mm punch helped to minimize mary visual neocortex (PV) had seeding in only n = 1/7 such occurrences. individuals at NFT stage III, and at NFT stage V n = 4/7 individuals scored positive. Cerebellar dentate nucleus and cerebellar cortex In two individuals, the cerebellar dentate nucleus (DN) and/or cerebellar cortex (CC) had seeding (Fig. 3a). Inter- Occasional/inconsistent seeding in 6 brain regions estingly, AT8 staining for the two positive CC regions We observed seeding and AT8-positive tau pathology in (cases 9 and 12) revealed AT8-positivity in the Bergmann some individuals at higher NFT stages (III and V) in 6 glia (as thorn-shaped astrocytes) and in their astrocytic additional brain regions, which is a new finding in some processes (Fig. 4j and k) rather than in neurons, which is cases. a novel ARTAG finding. In the DN of 4 cases, nerve cells plus axons were AT8-positive (Fig. 4i). Pontine gray Three individuals (n = 1 at NFT stage III and n = 2 at NFT stage V) scored positive for seeding in the pon- Internal capsule, anterior limb tine gray (PG). Of note, for case 7 (with positive seeding We observed seeding in the anterior limb of the internal of 0.79 in the PG), the dorsal raphe nucleus (DRN) was capsule in n = 1/7 case at NFT stage III, and in n = 4/7 punched accidentally instead of PG for the seeding assay cases at NFT stage V (Fig. 3a). AT8 staining in 5 cases (Fig. 4c). AT8 staining in n = 8/20 cases showed (mostly revealed positive axonal signals. The anterior portion of mild) AT8 changes in neurons and neurites/axons of the the inner capsule contains both axons of the frontopon- PG in n = 4 cases, of which n = 1 (case 16) had seeding tine projection and axons that originate in the mediodor- (Fig. 4b). sal complex of thalamus (MD) [42]. Neither the IC nor the OFC or MD were above threshold for seeding in NFT stage I, but this shifted at NFT stage III, and OFC and Inferior olivary nucleus MD displayed strong seeding at NFT stage V. 4 individuals (2 at NFT stage III and 2 at NFT stage V) had seeding in the inferior olivary nucleus (IO). AT8 staining in 9 cases was negative in neurons. However, Optic chiasm/tract three cases had tau-positive astrocytes consistent with Only n = 1/8 individual at NFT stage V (case 20) showed aging-related astrogliopathology (ARTAG) (Fig. 4f ), and positive seeding in the optic chiasm/tract (OC) (Fig. 3a). 6 cases had AT8-positive axons (Fig. 4d and e). Note that Additional tissue from this brain region was not available obtaining punches from the IO is technically challenging (See figure on next page.) Fig. 4 Phospho‑tau Histopathology (Part I). a The TRE of case 1 (female, 55 years, NFT I, Table 1) displayed a few AT8‑immunopositive neurons, but these were below the threshold for tau seeding (Fig. 3). With the exception of this case and case 3 (female, 50 years, NFT I, Table 1), the TRE of all 18 remaining individuals showed some degree of tau seeding activity (0.30–17.90, Fig. 3). b In case 16 (male, 58 years, NFT V, Table 1), moderate AT8 pathology (neuronal somata, axons) and low tau seeding (1.07) were detectable in the PG. c Punches mistakenly located in the dorsal raphe nucleus (DRN) of one individual (case 7, female, 93 years, NFT III, Table 1) showed AT8‑immunoreactive neuronal pathology combined with mild tau seeding (0.79). d Some IO punches (case 7), with no AT8‑positive cell bodies, contained isolated AT8‑positive axons and low tau seeding activity (0.60). e Framed area in d at higher magnification. Arrows indicate two AT8‑positive axons. f In the IO of case 7 (same individual as in c-e and g, h) we noted unexpected and marked aging‑related tau astrogliopathy, ARTAG (example in the framed area). g, h A single AT8‑positive neuron was seen in this punch from the GP accompanied by numerous AT8‑immunoreactive axons (see also Fig. 5f ), the latter possibly originating in the basal nucleus of Meynert (BN). Tau seeding in the GP of this case was moderate (2.65); notably, the BN of this individual also featured moderate tau seeding (5.46). I. The DN of case 9 (male, 72 years, NFT III, Table 1) had few AT8‑positive cell bodies (pretangles) and was sub ‑threshold for tau seeding (0.25). In cases at NFT stages I and III, only one (case 12, female, 81 years, NFT III, Table 1) had mild tau seeding in this region (0.60). j, k Notably, in the CC of case 12 (female, 81 years, NFT III, Table 1), the Bergmann glia (here, as thorn‑shaped astrocytes) and, in the molecular layer, their astrocytic processes were AT8‑immunopositive and also had tau seeding (1.39). By contrast, the neurons in the Purkinje layer were AT8‑negative and negative for tau seeding. Tau seeding (1.74) was detected in the CC of one additional individual (case 9, male, 72 years, NFT III, Table 1, Fig. 3) Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 12 of 19 Fig. 4 (See legend on previous page.) S topschinski et al. acta neuropathol commun (2021) 9:164 Page 13 of 19 to obtain reliable AT8 staining. Seeding in the OC might especially in closely adjacent anatomical regions, such as originate in axons from the DRN, the lower raphe nuclei, the IO and IRZ. However, smaller diameter punches also LC, amygdala, EC, superior colliculus, and/or the retina increased the variance in tau seeding detected, especially [42, 49]. in areas with scarce tau pathology, e.g., the IC. Further, given the small size of some brain regions examined we occasionally sampled adjacent sections, which may have ApoE status does not influence seeding in our study contributed to the variability. Finally, the high threshold ApoE genotyping was available for n = 16/20 cases. 8 car- to register a positive signal may have contributed to the ried at least one ApoE ε4 allele, and 2 were homozygous variability in regions with low numbers of tau-positive for ApoE ε4 (Table 1). We detected no significant influ - neurons. In prior work [28], adjacent punches from the ence of ApoE on the seeding in all cases comparing E4 TRE/EC correlated relatively well, which may be due carriers versus non-carriers. to its more easily defined structure, or more homoge - neously distributed pathology than many of the brain regions studied here. Other tauopathies can be excluded in our cohort None of the cases displayed the tauopathy changes spe- cific for CTE [37] or PSP [16, 33, 43, 45]. Seeding increases with neuropathological stage Some brain regions studied are not known to develop Discussion tau pathology in AD, yet we observed tau seed- We used an ultrasensitive biosensor assay to create a ing across all 25. Many regions included in this study comprehensive map of seeding in the AD brain, analyz- have not previously been studied for tau seeding. Our ing 25 brain regions (15 of which had not been analyzed predictions regarding seeding were based on prior previously) across 20 individuals with AD pathology neuropathological data and our knowledge of the neu- at NFT stages I, III, and V. We used a punch biopsy for roanatomical connections that determine input and precise sampling of regions of interest, and defined the output for these brain regions. Our predictions were positive seeding at 3 SD above the average of all negative accurate in many cases (Additional file 1: Table S2), e.g., samples tested, a threshold with relatively high specificity as predicted, we found early seeding in the transen- and low sensitivity. torhinal and entorhinal cortex; seeding at intermedi- ate stages in the superior temporal gyrus (STG) and the anterior cingulate cortex (ACC); and seeding at Variability within brain regions intermediate/late stages in the putamen and the medi- We observed variability between different punches from odorsal complex of thalamus (MD). These findings were the same brain regions, highlighting the importance in consistent with the tau progression pathway described future studies of sampling larger brain volumes or aver- in prior studies [4, 14, 28]. aging multiple biopsies. We used small punches with a diameter of 3–4 mm to increase sampling precision, (See figure on next page.) Fig. 5 Phospho‑tau Histopathology (Part II). a A 4 mm punch from the IO of case 16 (male, 58 years, NFT V ) included portions of the immediately adjacent intermediate zone (IRZ) with strongly AT8‑positive neurons and axons. In the IO itself, a single AT8‑immunoreactive axon was detectable and some perivascular ARTAG was present, but none of the neurons there were AT8‑immunoreactive. Tau seeding was below threshold (0.24). In at least two additional individuals (cases 12 and 17), prominent AT8‑positive neurons and axons in the IRZ accompanied by IO AT8‑negative neurons as well as perivascular ARTAG may have accounted for the tau seeding signals in IO punches (2.29 and 4.74). b Framed area in a at higher magnification showing AT8‑positive tau pathology at the punch edge. c A 3 mm punch with edges free of AT8 pathology from case 15 (female, 84 years, NFT V ), in which no portions of the IRZ were included. d Framed area in c at higher magnification displaying a clean punch edge and, directly beyond it, AT8‑positive pathology in the IRZ. Tau seeding activity in the IO of this case was below threshold (0.08). e AT8‑positive axon (arrows) in the substantia nigra, pars compacta (SNpc) of case 13 (female, 88 years, NFT III). Some tau seeding in the SNpc was present in 13/20 individuals (0.81–14.20), and was exceeded in the lower brainstem only by 15/20 cases in the LC (1.00–19.43). f Arrows point to AT8‑positive axons in the GP of case 7 (female, 93 years, NFT III) (see also Fig. 4g, h). g The highest propensity for tau seeding (33.77) was detected in the basolateral subnucleus of the amygdala (AMY ) of case 18 (female, 76 years, Stage V ), where severe diffuse neuronal AT8‑pathology was accompanied by some AT8‑immunopositive astrocytes (ARTAG). h, i Tau seeding activity in the LC first became more pronounced during NFT stages III and V, e.g., 11.96 in case 7 (female, 93 years, NFT III) (h); 19.43 in case 19 (male, 76 years, NFT III); and 5.71 in case 10 (male, 74 years, NFT III) in micrograph i. Arrow in i points to extraneuronal neuromelanin lying free in the neuropil after severe neuronal loss Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 14 of 19 Fig. 5 (See legend on previous page.) Tau seeding starts in the TRE both individuals 1 and 3 contained up to 9 AT8-positive The majority of individuals tested in our study scored neurons and scored negative for seeding, possibly due positive in the TRE, excepting 2 individuals at NFT to the low density of AT8-positive neurons and the rela- stage 1 (cases 1 and 3). Interestingly, TRE sections of tively high seeding threshold we used (Additional file 1: S topschinski et al. acta neuropathol commun (2021) 9:164 Page 15 of 19 Table 3 Unexpected seeding results Brain region Expectation Reference for expectation Findings 1‑ TRE Seeding in all cases Kaufmann et al. [28] Unexpected = > n = 2/6 cases at NFT stage I did not show seeding 4‑AMY Seeding starting at late NFT stages Braak et al. [4] Unexpected = > seeding in n = 1/6 NFT stage I case, in n = 6/7 NFT stage III cases and in all NFT stage V cases 6‑ TTG Seeding starting at late NFT stages ( V‑ VI) Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case; expectedly in n = 6/7 NFT stage V cases 12‑ GP Seeding not expected since it typically does not Braak and Del Tredic [6] Unexpected = > seeding in n = 3/7 NFT stage develop tau pathology (in contrast to Aβ plaques) III cases and in n = 6/7 NFT stage V cases in AD 15‑SN Seeding starting at late NFT stages ( V‑ VI) Braak et al. [4] Unexpected = > seeding seen earlier—in most NFT stage III cases (n = 6/7) and in all NFT stage V cases 18‑PG Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case and in n = 2/7 NFT stage V cases Note: The dorsal raphe nucleus (DRN) in case 7 was punched accidentally and likely explains the seeding in this sample 19‑IO Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 2/7 NFT stage III cases and in n = 2/7 NFT stage V cases Note: Some punches from IO were contaminated by over‑ lapping AT8‑positive neurons/axons in the IRZ 20‑ CC Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 2/7 NFT stage III cases 21‑DN Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case and in n = 1/7 NFT stage V case 22‑IC Seeding not expected Braak et al. [4] Unexpected = > seeding in n = 1/7 NFT stage III case and in n = 4/7 NFT stage V cases 24‑ OB Seeding starting at early/intermediate NFT stages Attems et al. [2] As expected, seeding in n = 4/7 NFT stage III cases; (II‑III) however, unexpected = > in only n = 5/6 NFT stage V cases 25‑ OC Seeding expected DeVos et al. [14] Unexpected = > seeding in only one individual (case 20) Summary of seeding results that were unexpected based on seeding results and/or neuropathology data from prior studies. For a complete list of all brain regions and expectations prior to the study refer to Additional file 1: Table S2. For unexpected results, AT8 IHC was performed as listed in Additional file 1: Table S1 Table S1). In other words, the number of AT8-positive the seeding assay reliably detects monomeric and oligo- structures (i.e., “severity” of pathology) in the TRE may meric tau seeds [39, 40], and we speculate that the pres- not account for the level of seed-competent tau. Prior ence of seed-competent tau monomer and oligomers can work from our group indicates that early tau pathology precede the presence of AT8-positive tangles. Studies in starts in the TRE/EC region [28]. Of note, Kaufman et al. the PS19 mouse model support these conclusions [25]. [28] did not separate the TRE and EC, as we did here. The differences in seeding between TRE and EC were subtle: Tau seeding occurs in unanticipated brain regions compared to the EC, the TRE had seeding in more indi- We detected tau seeding in brain regions not known to viduals (n = 4 vs. n = 2) at NFT stage I, and the average develop tau pathology, i.e. the GP, IC, PG, IO, and cer- seeding was slightly higher at NFT stage I. However, the ebellum. The seeding in these brain regions was low slightly earlier involvement of TRE is consistent with its overall, and neuropathological findings/AT8 staining and classification as NFT stage I [4]. seeding did not necessarily correlate with each other. In some regions, subsequent AT8 staining revealed the pres- Tau seeding precedes tau pathology ence of tau predominantly in astrocytes (see below). We Some brain regions show seeding significantly ear - also detected tau seeding in regions not known to exhibit lier than predicted based on neuropathological staging tau seeds, although observed to develop neurofibrillary (SNpc, AMY, TTG/Heschl’s gyrus). This confirms prior pathology during AD. These included the IRZ, DRN, observations that tau seeding can precede the develop- AMY, BN, and OB. The seeding was not always consist - ment of tau pathology as detected by immunohistochem- ent between connected anatomical regions within each istry [14, 20]. Work from our laboratory has shown that Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 16 of 19 Summary and conclusion individual case, since punches from different regions In the future, we will focus on selected brain regions were taken randomly from the right and left hemisphere, with notable and/or new results, such as the IRZ, AMY, and we did not intend to analyze direct anatomical con- BN, IC, GP, cerebellum, and OB. Moreover, an analy- nectivity between specific brain punches. Therefore, we sis of tau strains in all of the regions studied here and cannot speculate about anatomical pathways that could in the hypothalamus [4] would be informative. Impor- explain seeding in these samples. However, we note that tantly, although our data are limited by the number of the AT8-positive axons observed in the GP could belong individuals analyzed (n = 6–7 per NFT stage), we have to the BN. Seeding levels in the BN in all such cases were observed tau seeding in regions where AT8 staining always comparatively higher than in their respective GP is negative. This suggests that tau pathology extends punches (Fig. 3a). beyond the brain regions that are routinely included in AD staging, and is highly variable between individu- Aging-related tau astrogliopathy (ARTAG) als. Thus, immunohistochemistry and seeding are two In some brain regions with unexpected presence of tau fundamentally different and yet complementary meth - seeding (globus pallidus, inferior olivary nucleus, cer- ods for assessing tau pathology in AD. It remains to be ebellar cortex), we detected tau pathology in astrocytes determined whether a combination of these metrics or in Bergmann glia. It is not known how abnormal tau will help explain the diversity of AD presentation. in astrocytes accumulates [18]. ARTAG develops mainly, but not exclusively, in individuals over 60 years of age. Supplementary Information The two major cytomorphologies are thorn-shaped astro - The online version contains supplementary material available at https:// doi. cytes (TSAs) and granular or fuzzy tau immunoreactiv- org/ 10. 1186/ s40478‑ 021‑ 01255‑x. ity in astrocytic processes (GFA). TSAs occur in subpial, subependymal, or perivascular areas, as well as white Additional file 1: Tables S1 and S2. See separate online file. matter [31, 32]. It is now generally accepted that astro- Additional file 2: Figure S6 (a). Tau seeding map of NFT stage I. See glia either have no endogenous tau [22] or, very low lev- separate online files. MR imaging of the brain was performed in a healthy els [31]. It is unclear why tau accumulates in astrocytes, volunteer. Segmentation of regions of interest (ROIs) within the brain and whether it transfers between them. We hypothesize was performed semi‑automatically or by a board‑ certified neuroradiolo ‑ gist (F.F.Y.). The weighted average tau deposition from all subjects within that they phagocytize it from the surrounding intersti- each NFT stage group was then applied to each ROI. Brain regions were tial space, while tau seeds might originate in neurons included only if at least one subject within each NFT stage had positive [18]. Ferrer et al. have described seeding in AT8-positive tau seeding results. The ROIs were then visualized within a 3D projection of the volunteer’s brain. For clarity, 3D representation is limited to one astrocytes in cases displaying minimal intraneuronal tau hemisphere. A graded color scale from white to yellow to orange to red pathology without determining where seed competent (the same color coding as used in Fig. 3) indicates increasing amount of tau originates [18]. The pathophysiological significance seeding. Brain regions not sampled in this study are colored in grey. Note that many brains regions that are traditionally not included in AD pathol‑ of ARTAG for AD is unknown [31] and will require fur- ogy staging exhibit tau seeding, some of them at early NFT stages. ther studies. Our prior studies of tau strains indicate that Additional file 3: Figure S6 (b). Tau seeding map of NFT stage III. See some preferentially involve astrocytes [29, 47]. separate online files. MR imaging of the brain was performed in a healthy volunteer. Segmentation of regions of interest (ROIs) within the brain was performed semi‑automatically or by a board‑ certified neuroradiolo ‑ Tau seeding in the cerebellum and tau pathology gist (F.F.Y.). The weighted average tau deposition from all subjects within each NFT stage group was then applied to each ROI. Brain regions were in Bergmann glia included only if at least one subject within each NFT stage had positive We predicted that the cerebellar dentate nucleus (DN) tau seeding results. The ROIs were then visualized within a 3D projection and the cerebellar cortex (CC) would not exhibit seeding, of the volunteer’s brain. For clarity, 3D representation is limited to one hemisphere. A graded color scale from white to yellow to orange to red however, for each region, n = 2/20 cases did. We did not (the same color coding as used in Fig. 3) indicates increasing amount of observe AT8-positive tau pathology in the somatoden- seeding. Brain regions not sampled in this study are colored in grey. Note dritic or axonal compartments of cerebellar neurons, e.g., that many brains regions that are traditionally not included in AD pathol‑ ogy staging exhibit tau seeding, some of them at early NFT stages. granule cells, Purkinje cells, or stellate/basket cells. Nota- Additional file 4: Figure S6 (c). Tau seeding map of NFT stage V. See bly, AT8 staining for the 2 seeding-positive CC regions separate online files. MR imaging of the brain was performed in a healthy revealed AT8 staining of Bergmann glia rather than volunteer. Segmentation of regions of interest (ROIs) within the brain neurons and axons (Fig. 4j, k) (see remarks under the was performed semi‑automatically or by a board‑ certified neuroradiolo ‑ gist (F.F.Y.). The weighted average tau deposition from all subjects within previous heading ARTAG). We previously observed low each NFT stage group was then applied to each ROI. Brain regions were tau seeding in the cerebellum at late stages [20] but tau included only if at least one subject within each NFT stage had positive pathology in Bergmann glia has not been described. The tau seeding results. The ROIs were then visualized within a 3D projection of the volunteer’s brain. For clarity, 3D representation is limited to one findings raise the possibility of a sub-type of AD, with a hemisphere. A graded color scale from white to yellow to orange to red tau strain that preferentially affects Bergmann glia cells. (the same color coding as used in Fig. 3) indicates increasing amount of S topschinski et al. acta neuropathol commun (2021) 9:164 Page 17 of 19 4. Braak H, Braak E (1991) Neuropathological stageing of Alzheimer‑related seeding. Brain regions not sampled in this study are colored in grey. Note changes. Acta Neuropathol 82:239–259. https:// doi. org/ 10. 1007/ BF003 that many brains regions that are traditionally not included in AD pathol‑ ogy staging exhibit tau seeding, some of them at early NFT stages. 5. Braak H, Braak E, Grundke‑Iqbal I, Iqbal K (1986) Occurrence of neuropil threads in the senile human brain and in Alzheimer’s disease: a third location of paired helical filaments outside of neurofibrillary tangles and Acknowledgements neuritic plaques. Neurosci Lett 65:351–355. https:// doi. org/ 10. 1016/ 0304‑ We acknowledge support from the Aging Minds Foundation (B.E.S., M.I.D.), the 3940(86) 90288‑0 Cure Alzheimer’s Foundation (B.E.S., M.I.D.), the Berry Cox Foundation (B.E.S, 6. Braak H, Del Trecidi K (2015) Neuroanatomy and pathology of sporadic M.I.D.), the King Foundation (B.E.S.), the Chan‑Zuckerberg Initiative (M.I.D.), the Alzheimer’s disease. Adv Anat Embryol Cell Biol 215:1–162 NIH/NIA RF1AG059689 (M.I.D.), the Hans & Ilse Breuer Foundation, Frank‑ 7. Braak H, Del Tredici K (2014) Are cases with tau pathology occurring furt am Main, Germany (H.B., K.D.T ), Ms. Simone Feldengut (silver staining, in the absence of Abeta deposits part of the AD‑related pathologi‑ immunohistochemistry), and Mr. David Ewert (Figure 4, 5 layouts) for skillful cal process? Acta Neuropathol 128:767–772. https:// doi. org/ 10. 1007/ technical assistance. s00401‑ 014‑ 1356‑1 8. Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN, Braak E (2003) Authors’ contributions Staging of brain pathology related to sporadic Parkinson’s disease. Neuro‑ B.E.S designed and performed all cell culture and flow cytometry experiments. biol Aging 24:197–211. https:// doi. org/ 10. 1016/ s0197‑ 4580(02) 00065‑9 S.E. assisted with animal tissue collection and preparation of animal tissue. 9. Bras J, Guerreiro R, Darwent L, Parkkinen L, Ansorge O, Escott‑Price V, K.D.T. and H.B. performed all human tissue collection, IHC and neuropatho‑ Hernandez DG, Nalls MA, Clark LN, Honig LS et al (2014) Genetic analysis logical staging. E.G. performed APOE genotyping. F.F.Y. generated the 3D tau implicates APOE, SNCA and suggests lysosomal dysfunction in the seeding maps. M.I.D. assisted with the design and interpretation of all flow etiology of dementia with Lewy bodies. Hum Mol Genet 23:6139–6146. cytometry experiments. All authors assisted in the writing and figure prepara‑ https:// doi. org/ 10. 1093/ hmg/ ddu334 tion for this manuscript. All authors read and approved the final manuscript. 10. Calafate S, Buist A, Miskiewicz K, Vijayan V, Daneels G, de Strooper B, de Wit J, Verstreken P, Moechars D (2015) Synaptic contacts enhance cell‑to ‑ Availability of data and materials cell tau pathology propagation. Cell Rep 11:1176–1183. https:// doi. org/ The datasets used and/or analyzed during the current study are available from 10. 1016/j. celrep. 2015. 04. 043 the corresponding author by reasonable request. 11. Clavaguera F, Goedert M, Tolnay M (2010) Induction and spreading of tau pathology in a mouse model of Alzheimer’s disease. Med Sci (Paris) 26:121–124. https:// doi. org/ 10. 1051/ medsci/ 20102 62121 Declarations 12. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak‑ Vance MA (1993) Gene dose of apoli‑ Ethics approval and consent to participate poprotein E type 4 allele and the risk of Alzheimer’s disease in late onset All animal maintenance and experiments adhered to the animal care and use families. Science 261:921–923. https:// doi. org/ 10. 1126/ scien ce. 83464 43 protocols of the University of Texas Southwestern Medical Center. The autopsy 13. Crary JF, Trojanowski JQ, Schneider JA, Abisambra JF, Abner EL, brains used for this study were obtained in compliance with Ulm University Alafuzoff I, Arnold SE, Attems J, Beach TG, Bigio EH et al (2014) Primary ethics committee guidelines as well as German federal and state law govern‑ age‑related tauopathy (PART ): a common pathology associated with ing human tissue usage. human aging. Acta Neuropathol 128:755–766. https:// doi. org/ 10. 1007/ s00401‑ 014‑ 1349‑0 Consent for publication 14. DeVos SL, Corjuc BT, Oakley DH, Nobuhara CK, Bannon RN, Chase A, Not applicable. Commins C, Gonzalez JA, Dooley PM, Frosch MP et al (2018) Synaptic tau seeding precedes tau pathology in human Alzheimer’s disease brain. Competing interests Front Neurosci 12:267. https:// doi. org/ 10. 3389/ fnins. 2018. 00267 MID is co‑ developer of an anti‑tau antibody currently in clinical trials (C2N 15. Duyckaerts C, Braak H, Brion JP, Buee L, Del Tredici K, Goedert M, Halliday 8E12 [NCT02494024]). The remaining authors declare that they have no G, Neumann M, Spillantini MG, Tolnay M et al (2015) PART is part of Alz‑ competing interests. heimer disease. Acta Neuropathol 129:749–756. https:// doi. org/ 10. 1007/ s00401‑ 015‑ 1390‑7 Author details 16. Ebashi M, Ito Y, Uematsu M, Nakamura A, Hirokawa K, Kamei S, Uchihara Center for Alzheimer’s and Neurodegenerative Diseases, Peter O’Donnell Jr. T (2019) How to demix Alzheimer‑type and PSP ‑type tau lesions out of Brain Institute, NL10.120, University of Texas Southwestern Medical Center, their mixture ‑hybrid approach to dissect comorbidity. Acta Neuropathol 6000 Harry Hines Blvd., Dallas, TX 75390, USA. Clinical Neuroanatomy Section/ Commun 7:71. https:// doi. org/ 10. 1186/ s40478‑ 019‑ 0708‑4 Department of Neurology, Center for Biomedical Research, University of Ulm, 17. Elobeid A, Libard S, Leino M, Popova SN, Alafuzoff I (2016) Altered pro ‑ Ulm, Germany. Institute of Clinical Neuroanatomy, J. W. Goethe University, teins in the aging brain. J Neuropathol Exp Neurol 75:316–325. https:// Frankfurt am Main, Germany. Department for Radiology, Neuroradiology Divi‑ doi. org/ 10. 1093/ jnen/ nlw002 sion, University of Texas Southwestern Medical Center, Dallas, TX, USA. 18. Ferrer I, Garcia MA, Gonzalez IL, Lucena DD, Villalonga AR, Tech MC, Llor‑ ens F, Garcia‑Esparcia P, Martinez‑Maldonado A, Mendez MF et al (2018) Received: 3 September 2021 Accepted: 3 September 2021 Aging‑related tau astrogliopathy (ARTAG): not only tau phosphorylation in astrocytes. Brain Pathol 28:965–985. https:// doi. org/ 10. 1111/ bpa. 12593 19. Furman JL, Holmes BB, Diamond MI (2015) Sensitive Detection of Pro‑ teopathic Seeding Activity with FRET Flow Cytometry. J Vis Exp: e53205 https:// doi. org/ 10. 3791/ 53205 References 20. Furman JL, Vaquer‑Alicea J, White CL 3rd, Cairns NJ, Nelson PT, Diamond 1. Ashford JW, Soultanian NS, Zhang SX, Geddes JW (1998) Neuropil threads MI (2017) Widespread tau seeding activity at early Braak stages. Acta are collinear with MAP2 immunostaining in neuronal dendrites of Alz‑ Neuropathol 133:91–100. https:// doi. org/ 10. 1007/ s00401‑ 016‑ 1644‑z heimer brain. J Neuropathol Exp Neurol 57:972–978. https:// doi. org/ 10. 21. Ghebremedhin E, Braak H, Braak E, Sahm J (1998) Improved method 1097/ 00005 072‑ 19981 0000‑ 00009 facilitates reliable APOE genotyping of genomic DNA extracted from for‑ 2. Attems J, Lintner F, Jellinger KA (2005) Olfactory involvement in aging maldehyde‑fixed pathology specimens. J Neurosci Methods 79:229–231. and Alzheimer’s disease: an autopsy study. J Alzheimers Dis 7:149–157; https:// doi. org/ 10. 1016/ s0165‑ 0270(97) 00197‑0 discussion 173–180. https:// doi. org/ 10. 3233/ jad‑ 2005‑ 7208 22. Goedert M, Spillantini MG, Potier MC, Ulrich J, Crowther RA (1989) Clon‑ 3. Braak H, Alafuzoff I, Arzberger T, Kretzschmar H, Del Tredici K (2006) ing and sequencing of the cDNA encoding an isoform of microtubule‑ Staging of Alzheimer disease‑associated neurofibrillary pathology associated protein tau containing four tandem repeats: differential using paraffin sections and immunocytochemistry. Acta Neuropathol expression of tau protein mRNAs in human brain. EMBO J 8:393–399 112:389–404. https:// doi. org/ 10. 1007/ s00401‑ 006‑ 0127‑z Stopschinski et al. acta neuropathol commun (2021) 9:164 Page 18 of 19 23. Hitt BD, Vaquer‑Alicea J, Manon VA, Beaver JD, Kashmer OM, Garcia JN, 40. Mirbaha H, Holmes BB, Sanders DW, Bieschke J, Diamond MI (2015) Tau Diamond MI (2021) Ultrasensitive tau biosensor cells detect no seeding trimers are the minimal propagation unit spontaneously internalized to in Alzheimer’s disease CSF. Acta Neuropathol Commun 9:99. https:// doi. seed intracellular aggregation. J Biol Chem 290:14893–14903. https:// doi. org/ 10. 1186/ s40478‑ 021‑ 01185‑8org/ 10. 1074/ jbc. M115. 652693 24. Hoenig MC, Bischof GN, Seemiller J, Hammes J, Kukolja J, Onur OA, Jessen 41. Nelson PT, Alafuzoff I, Bigio EH, Bouras C, Braak H, Cairns NJ, Castellani F, Fliessbach K, Neumaier B, Fink GR et al (2018) Networks of tau distribu‑ RJ, Crain BJ, Davies P, Del Tredici K et al (2012) Correlation of Alzheimer tion in Alzheimer’s disease. Brain 141:568–581. https:// doi. org/ 10. 1093/ disease neuropathologic changes with cognitive status: a review of the brain/ awx353 literature. J Neuropathol Exp Neurol 71:362–381. https:// doi. org/ 10. 1097/ 25. Holmes BB, Furman JL, Mahan TE, Yamasaki TR, Mirbaha H, Eades WC, NEN. 0b013 e3182 5018f7 Belaygorod L, Cairns NJ, Holtzman DM, Diamond MI (2014) Proteo‑ 42. Nieuwenhuys R, Voogd J, Huijzen C (2008) The human central nervous pathic tau seeding predicts tauopathy in vivo. Proc Natl Acad Sci U S A system: a synopsis and atlas, 4th edn. Springer, Berlin 111:E4376‑4385. https:// doi. org/ 10. 1073/ pnas. 14116 49111 43. Nogami A, Yamazaki M, Saito Y, Hatsuta H, Sakiyama Y, Takao M, Kimura 26. Hyman BT, Phelps CH, Beach TG, Bigio EH, Cairns NJ, Carrillo MC, Dickson K, Murayama S (2015) Early stage of progressive supranuclear palsy: a DW, Duyckaerts C, Frosch MP, Masliah E et al (2012) National institute on neuropathological study of 324 consecutive autopsy cases. J Nippon Aging‑Alzheimer’s association guidelines for the neuropathologic assess‑ Med Sch 82:266–273. https:// doi. org/ 10. 1272/ jnms. 82. 266 ment of Alzheimer’s disease. Alzheimers Dement 8:1–13. https:// doi. org/ 44. Ossenkoppele R, Schonhaut DR, Scholl M, Lockhart SN, Ayakta N, Baker 10. 1016/j. jalz. 2011. 10. 007 SL, O’Neil JP, Janabi M, Lazaris A, Cantwell A et al (2016) Tau PET patterns 27. Ishiki A, Okamura N, Furukawa K, Furumoto S, Harada R, Tomita N, Hiraoka mirror clinical and neuroanatomical variability in Alzheimer’s disease. K, Watanuki S, Ishikawa Y, Tago T et al (2015) Longitudinal assessment of Brain 139:1551–1567. https:// doi. org/ 10. 1093/ brain/ aww027 tau pathology in patients with Alzheimer’s disease using [18F]THK‑5117 45. Rüb U, Del Tredici K, Schultz C, de Vos RA, Jansen Steur EN, Arai K, Braak positron emission tomography. PLoS ONE 10:e0140311. https:// doi. org/ H (2002) Progressive supranuclear palsy: neuronal and glial cytoskeletal 10. 1371/ journ al. pone. 01403 11 pathology in the higher order processing autonomic nuclei of the lower 28. Kaufman SK, Del Tredici K, Thomas TL, Braak H, Diamond MI (2018) Tau brainstem. Neuropathol Appl Neurobiol 28:12–22. https:// doi. org/ 10. seeding activity begins in the transentorhinal/entorhinal regions and 1046/j. 0305‑ 1846. 2001. 00374.x anticipates phospho‑tau pathology in Alzheimer’s disease and PART. Acta 46. Rüb U, Del Tredici K, Schultz C, Thal DR, Braak E, Braak H (2001) The Neuropathol 136:57–67. https:// doi. org/ 10. 1007/ s00401‑ 018‑ 1855‑6 autonomic higher order processing nuclei of the lower brain stem are 29. Kaufman SK, Sanders DW, Thomas TL, Ruchinskas AJ, Vaquer‑Alicea among the early targets of the Alzheimer’s disease‑related cytoskeletal J, Sharma AM, Miller TM, Diamond MI (2016) Tau prion strains dictate pathology. Acta Neuropathol 101:555–564. https:// doi. org/ 10. 1007/ s0040 patterns of cell pathology, progression rate, and regional vulnerability 10000 320 In Vivo. Neuron 92:796–812. https:// doi. org/ 10. 1016/j. neuron. 2016. 09. 055 47. Sanders DW, Kaufman SK, DeVos SL, Sharma AM, Mirbaha H, Li A, Barker 30. Kaufman SK, Thomas TL, Del Tredici K, Braak H, Diamond MI (2017) Char‑ SJ, Foley AC, Thorpe JR, Serpell LC et al (2014) Distinct tau prion strains acterization of tau prion seeding activity and strains from formaldehyde‑ propagate in cells and mice and define different tauopathies. Neuron fixed tissue. Acta Neuropathol Commun 5:41. https:// doi. org/ 10. 1186/ 82:1271–1288. https:// doi. org/ 10. 1016/j. neuron. 2014. 04. 047 s40478‑ 017‑ 0442‑8 48. Sanders DW, Kaufman SK, Holmes BB, Diamond MI (2016) Prions and 31. Kovacs GG (2020) Astroglia and tau: new perspectives. Front Aging Neu‑ Protein Assemblies that Convey Biological Information in Health and rosci 12:96. https:// doi. org/ 10. 3389/ fnagi. 2020. 00096 Disease. Neuron 89:433–448. https:// doi. org/ 10. 1016/j. neuron. 2016. 01. 32. Kovacs GG, Ferrer I, Grinberg LT, Alafuzoff I, Attems J, Budka H, Cairns NJ, 026 Crary JF, Duyckaerts C, Ghetti B et al (2016) Aging‑related tau astro ‑ 49. Schön C, Hoffmann NA, Ochs SM, Burgold S, Filser S, Steinbach S, Seeliger gliopathy (ARTAG): harmonized evaluation strategy. Acta Neuropathol MW, Arzberger T, Goedert M, Kretzschmar HA et al (2012) Long‑term 131:87–102. https:// doi. org/ 10. 1007/ s00401‑ 015‑ 1509‑x in vivo imaging of fibrillar tau in the retina of P301S transgenic mice. PLoS 33. Kovacs GG, Lukic MJ, Irwin DJ, Arzberger T, Respondek G, Lee EB, Cough‑ ONE 7:e53547. https:// doi. org/ 10. 1371/ journ al. pone. 00535 47 lin D, Giese A, Grossman M, Kurz C et al (2020) Distribution patterns 50. Singleton AB, Wharton A, O’Brien KK, Walker MP, McKeith IG, Ballard of tau pathology in progressive supranuclear palsy. Acta Neuropathol CG, O’Brien J, Perry RH, Ince PG, Edwardson JA et al (2002) Clinical and 140:99–119. https:// doi. org/ 10. 1007/ s00401‑ 020‑ 02158‑2 neuropathological correlates of apolipoprotein E genotype in dementia 34. Kovacs GG, Milenkovic I, Wohrer A, Hoftberger R, Gelpi E, Haberler C, with Lewy bodies. Dement Geriatr Cogn Disord 14:167–175. https:// doi. Honigschnabl S, Reiner‑ Concin A, Heinzl H, Jungwirth S et al (2013) org/ 10. 1159/ 00006 6022 Non‑Alzheimer neurodegenerative pathologies and their combina‑ 51. Sun A, Nguyen XV, Bing G (2002) Comparative analysis of an improved tions are more frequent than commonly believed in the elderly brain: thioflavin‑s stain, Gallyas silver stain, and immunohistochemistry for a community‑based autopsy series. Acta Neuropathol 126:365–384. neurofibrillary tangle demonstration on the same sections. J Histochem https:// doi. org/ 10. 1007/ s00401‑ 013‑ 1157‑y Cytochem 50:463–472. https:// doi. org/ 10. 1177/ 00221 55402 05000 403 35. Lee VM, Goedert M, Trojanowski JQ (2001) Neurodegenerative tauopa‑ 52. Tan RH, Shepherd CE, Kril JJ, McCann H, McGeachie A, McGinley C, Affleck thies. Annu Rev Neurosci 24:1121–1159. https:// doi. org/ 10. 1146/ annur ev. A, Halliday GM (2013) Classification of FTLD ‑ TDP cases into pathological neuro. 24.1. 1121 subtypes using antibodies against phosphorylated and non‑phospho ‑ 36. Lowe VJ, Wiste HJ, Senjem ML, Weigand SD, Therneau TM, Boeve BF, rylated TDP43. Acta Neuropathol Commun 1:33. https:// doi. org/ 10. 1186/ Josephs KA, Fang P, Pandey MK, Murray ME et al (2018) Widespread brain 2051‑ 5960‑1‑ 33 tau and its association with ageing, braak stage and Alzheimer’s demen‑ 53. Tsuang D, Leverenz JB, Lopez OL, Hamilton RL, Bennett DA, Schneider tia. Brain 141:271–287. https:// doi. org/ 10. 1093/ brain/ awx320 JA, Buchman AS, Larson EB, Crane PK, Kaye JA et al (2013) APOE epsilon4 37. McKee AC, Cairns NJ, Dickson DW, Folkerth RD, Keene CD, Litvan I, Perl DP, increases risk for dementia in pure synucleinopathies. JAMA Neurol Stein TD, Vonsattel JP, Stewart W et al (2016) The first NINDS/NIBIB con‑ 70:223–228. https:// doi. org/ 10. 1001/ jaman eurol. 2013. 600 sensus meeting to define neuropathological criteria for the diagnosis of 54. Uchihara T (2007) Silver diagnosis in neuropathology: principles, practice chronic traumatic encephalopathy. Acta Neuropathol 131:75–86. https:// and revised interpretation. Acta Neuropathol 113:483–499. https:// doi. doi. org/ 10. 1007/ s00401‑ 015‑ 1515‑zorg/ 10. 1007/ s00401‑ 007‑ 0200‑2 38. Mercken M, Vandermeeren M, Lubke U, Six J, Boons J, Van de Voorde 55. Wennberg AM, Tosakulwong N, Lesnick TG, Murray ME, Whitwell JL, A, Martin JJ, Gheuens J (1992) Monoclonal antibodies with selective Liesinger AM, Petrucelli L, Boeve BF, Parisi JE, Knopman DS et al (2018) specificity for Alzheimer Tau are directed against phosphatase ‑sensitive Association of Apolipoprotein E epsilon4 With Transactive Response epitopes. Acta Neuropathol 84:265–272. https:// doi. org/ 10. 1007/ BF002 DNA‑Binding Protein 43. JAMA Neurol 75:1347–1354. https:// doi. org/ 10. 278191001/ jaman eurol. 2018. 3139 39. Mirbaha H, Chen D, Morazova OA, Ruff KM, Sharma AM, Liu X, Goodarzi 56. Xia C, Makaretz SJ, Caso C, McGinnis S, Gomperts SN, Sepulcre J, M, Pappu RV, Colby DW, Mirzaei H et al (2018) Inert and seed‑ competent Gomez‑Isla T, Hyman BT, Schultz A, Vasdev N et al (2017) Association tau monomers suggest structural origins of aggregation. Elife. https:// doi. of in vivo [18F]AV‑1451 tau PET imaging results with cortical atrophy org/ 10. 7554/ eLife. 36584 S topschinski et al. acta neuropathol commun (2021) 9:164 Page 19 of 19 and symptoms in typical and atypical Alzheimer disease. JAMA Neurol Publisher’s Note 74:427–436. https:// doi. org/ 10. 1001/ jaman eurol. 2016. 5755 Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 57. Yoshiyama Y, Higuchi M, Zhang B, Huang SM, Iwata N, Saido TC, Maeda lished maps and institutional affiliations. J, Suhara T, Trojanowski JQ, Lee VM (2007) Synapse loss and microglial activation precede tangles in a P301S tauopathy mouse model. Neuron 53:337–351. https:// doi. org/ 10. 1016/j. neuron. 2007. 01. 010 Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your ﬁeld rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Acta Neuropathologica Communications – Springer Journals

Published: Oct 11, 2021

Keywords: Alzheimer’s disease; AT8; FRET biosensor; Neurofibrillary tangles; Prion propagation; Tau seeding; NFT staging

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Anatomic survey of seeding in Alzheimer’s disease brains reveals unexpected patterns

Anatomic survey of seeding in Alzheimer’s disease brains reveals unexpected patterns

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Anatomic survey of seeding in Alzheimer’s disease brains reveals unexpected patterns

Anatomic survey of seeding in Alzheimer’s disease brains reveals unexpected patterns

References (58)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies