Development and Characterization of Mouse-Specific Anti-Tau Monoclonal Antibodies: Relevance for Analysis of Murine Tau in Cerebrospinal Fluid

Abstract

Background:

Clearance of tau seeds by immunization with tau antibodies is currently evaluated as therapeutic strategy to block the spreading of tau pathology in Alzheimer’s disease and other tauopathies. Preclinical evaluation of passive immunotherapy is performed in different cellular culture systems and in wild-type and human tau transgenic mouse models. Depending on the preclinical model used, tau seeds or induced aggregates can either be of mouse, human or mixed origin.

Objective:

We aimed to develop human and mouse tau-specific antibodies to discriminate between the endogenous tau and the introduced form in preclinical models.

Methods:

Using hybridoma technology, we developed human and mouse tau-specific antibodies that were then used to develop several assays to specifically detect mouse tau.

Results:

Four antibodies, mTau3, mTau5, mTau8, and mTau9, with a high degree of specificity for mouse tau were identified. Additionally, their potential application in highly sensitive immunoassays to measure tau in mouse brain homogenate and cerebrospinal fluid is illustrated, as well as their application for specific endogenous mouse tau aggregation detection.

Conclusion:

The antibodies reported here can be very important tools to better interpret the results obtained from different model systems as well as to study the role of endogenous tau in tau aggregation and pathology observed in the diverse mouse models available.

Keywords

Antibodies assay development cerebrospinal fluid murine tau tau aggregation

INTRODUCTION

Tau is a microtubule-associated protein expressed in neuronal cells of the central nervous system (CNS) [1]. In the CNS, this protein can be present in 6 different isoforms which differ in the number of microtubule-binding repeats and N-terminal inserts [2]. Its main known functions are the assembly and stabilization of microtubules. However, there is evidence that tau can also exert other functions such as DNA protection and regulation of axonal transport and play a role in cellular processes such as neurogenesis and synaptic plasticity [3 –7].

Tau protein has been involved in several pathologies generally termed tauopathies, which include frontotemporal dementia with parkinsonism-17 (FTDP-17), Pick’s disease, progressive supranuclear palsy, corticobasal degeneration, argyrophilic grain disease, and Alzheimer’s disease [8]. They all share symptomatology (dementia), yet each disorder has its particular pathological features. In these disorders, tau usually presents abnormal patterns of post-translational modifications and forms aggregates within the cells that lead to the formation of neurofibrillary tangles, ultimately resulting in neuronal death [9, 10]. The presence and distribution of these tangles has been shown to correlate with cognitive decline in the different tauopathies, demonstrating the neuropathological potential of tau [11].

To study tau pathology, cellular and animal models are broadly used. The most common models are based on genetically modified mice to (over)express mutant forms of human tau [12, 13]. Nevertheless, with the exception of knock-in models, the mice still express endogenous murine tau. Most studies using mouse models tend to overlook the contribution that the endogenous tau may have in development of the pathology and since murine and human tau share a high degree of homology it makes it difficult to distinguish one from the other [14]. Additionally, seeds that are used in cellular and mouse seeding models, are mostly of human nature [15 –18] and discrimination of these molecules from endogenous protein is necessary for proper interpretation.

Antibodies play a central role in research as they are the basis of many techniques. The lack of mouse-specificity in commercially available antibodies makes it difficult to be able to separate and evaluate what happens to endogenous tau and how involved this protein is in the observed pathology.

To overcome this hurdle in the study of tau pathology, we have developed and characterized anti-tau antibodies that specifically recognize either murine or human tau. Furthermore, we demonstrate how these antibodies can be used to evaluate the presence of murine tau in cerebrospinal fluid (CSF) from the P301S model of tau pathology [19], as well as the presence of mouse tau aggregates in brain homogenates and mouse primary neurons.

MATERIALS AND METHODS

All in vivo experiments were conducted in strict accordance to the guidelines of the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC), with the European Communities Council Directive of 24 November 1986 (86/609/EEC) and with protocols approved by the local Institutional Animal and Use Ethical Committee.

Immunization of Balb/c mice and clone selection

Balb/c mice were injected intraperitoneally and subcutaneously with 75μg of recombinant full-length murine tau (mTau) or 100μg of recombinant full-length human tau (hTau) dissolved in Freund’s Complete Adjuvant (Sigma, F-5881). After 1 month a booster was given intraperitoneally with 75μg mTau or 100μg hTau in Freund’s Incomplete Adjuvant (Sigma, F-5506). Ten days after this booster, blood was collected via tail bleeding and antibody titers for mTau or hTau were determined. Mice with the highest titers received a second and final booster of mTau or hTau in phosphate-buffered saline (PBS) injected intraperitoneal 2 weeks after the first bleeding.

Hybridomas were then generated by fusing harvested spleen cells with myeloma cells as described in [20].

Selection of the clones

The obtained hybridomas were seeded in 96-well plates. 10 days later the hybridoma medium was tested in a screening ELISA and all clones of a positive well were again plated at 2 cells/well and grown for one week. A second screening ELISA was done, and the positive clones were kept for monoclonal antibody production.

All hybridomas were grown in Dulbecco’s modified Eagle’s medium (Thermo Fisher Scientific, 11960-044) supplemented with 10% fetal calf serum (Biowest, S1810-500), Hybridoma Fusion Cloning Supplement (Merck, 11363735001), HAT (Merck, H-0262), 1 mM sodium pyruvate (Merck, S-8636), 2 mM L-glutamine (Sigma, G3126), penicillin (100 U/ml), and streptomycin (50μg/ml) (Merck, P4333).

Purification of mouse IgG monoclonal antibodies

Hybridoma cells were grown in conventional Dulbecco’s medium (Life Technologies, 41965-039) supplemented with 10% serum, HT (Sigma, H-0137) and Hybridoma Fusion Cloning Supplement (Sigma, 11363735001). Cell growth was monitored using a Countess^® Automated Cell Counter (Invitrogen, MA, USA) until the viable cell density reached 5–8×10⁵ cells/mL. Cells were then subcultured to a viable cell density of 1×10⁵ cells/mL in fresh prewarmed Hybridoma-SF Medium (Life Technologies, 12045-084). After an incubation of 5 days at 37°C, cell viability was checked and when the viability was lower than 50%, cells were centrifuged for 10 min at 1000 g and consequently the supernatant was filtered through a 0.22μm filter (Millipore, SLGSM33SS). Next, 1L of filtered cell supernatant was run with a peristaltic pump over a HiTrap Prot G HP column (GE Healthcare, 17-0404-01) that was equilibrated with 5 column volumes of binding buffer (PBS) immediately followed by application of the supernatant. After washing with binding buffer, the antibodies were eluted with elution buffer (0.1 M Glycine-HCl pH 2.7) and collected in a 10% (v/v) 100 mM Tris pH 9 buffer. Fractions that were protein positive in Nanodrop 8000 (Thermo Fisher Scientific, Waltham, MA) were pooled. Dialysis was done against PBS in a Slide-A-Lyzer MWCO 10000 (Thermo Fisher Scientific, 66380). Concentration was determined by measuring the optical density (OD) at 280 nm and the antibody was filter sterilized, aliquoted and stored at –80°C.

Epitope mapping

To define the epitope of the antibodies, a library of peptides (20-mers with an overlap of 18 amino acids), covering the tau 441 sequence, was screened by Pepscan using its proprietary Chemically Linked Peptides on Scaffolds (CLIPS) technology [21, 22] (Pepscan Presto B.V., Lelystad, the Netherlands). To define mouse versus human specificity, a focused peptide array comprising the region 105–152 in human 2N4R tau and the corresponding region in mouse tau, was used. The peptides were incubated overnight at 4°C with primary antibody solution, followed by a 1 h incubation at 25°C with peroxidase conjugated anti-mouse antibody at 1/1000 dilution and another 1 h incubation with peroxidase substrate 2,2’-azino-di-3-ethylbenzthiazoline sulfonate (ABTS) and 20μl/mL of 3% H₂O₂. All incubation steps were proceeded by washing steps. Color development was measured and quantified with a charge coupled device-camera and an image processing system.

Indirect ELISA

Nunc MaxiSorptrademark flat-bottom 96-well plates (Thermo Fisher Scientific, 430341) were coated either with 10 ng/ml of full-length recombinant mouse or human tau in coating buffer (10 mM NaCl, 10 mM Tris-HCl, pH = 8.6), and left to incubate overnight at 4°C. After overnight incubation the plates were washed 5× with wash buffer (0.05% Tween-20 in PBS) followed by a 2 h incubation at room temperature (RT) in blocking buffer (0.1% casein in PBS). After another wash, primary antibody (mTau2, mTau3, mTau5, mTau8, mTau9, or HT7; Thermo Fisher Scientific, MN1000) was added in three-fold serial dilutions. After 2 h incubation at RT, the plates were washed and the secondary antibody goat anti-mouse IgG-HRP (Bio-Rad, 170-6516), diluted 1:2500 in blocking buffer, was incubated for 2 h at RT. Following incubation, the plates were washed and 50μl TMB working solution (hydrogen peroxide and TMB (3,3^′,5,5^′-tetramethylbenzidine) solution 1:9; Bio-Rad, 1721067) was added the wells. The enzymatic reaction was stopped with 50μl of 2N H₂SO₄. Plates were read immediately on EnVision^® 2102 Multilabel plate reader (PerkinElmer, Waltham, MA) and data was analyzed.

For hybridoma screening, hybridoma supernatant was added instead of primary antibody and incubations were done at 37°C. For detection anti-mouse IgG biotinylated (Amersham, RPN-1177) 1/2000 and streptavidin-HRPO (Amersham, RPN-1231) 1/2000 were used.

Meso Scale Discovery (MSD)

96-well Multi-Array^® plates (Meso Scale Discovery, L15XA-3) were incubated overnight at 4°C with each of the mouse antibodies diluted in PBS at a final concentration of 1μg/ml. After overnight incubation plates were washed 5 times with wash buffer (0.05% Tween in PBS) and then incubated with blocking buffer (0.1% casein in PBS) for 1.5-2 h at RT with agitation at 400 rpm. Next, the plates were washed again and brain homogenates from different species were loaded into the plates, in three-fold serial dilutions in blocking buffer. Plates were then sealed and incubated overnight at 4°C. Following incubation, the plates were washed and incubated with the detection antibody (sulfo-labelled PT82) diluted 1:1000 in blocking buffer for 1.5-2 h at RT with agitation at 400 rpm. After incubation, the plates are washed and 150μl of MSD Read Buffer T with surfactant (Meso Scale Discovery, R92TC) 2× diluted in distilled water was added to each well. Plates were immediately read using the MSD SECTOR Imager 6000 (Meso Scale Discovery, Gaithersburg, MD).

Immunohistochemistry

Sagittal 5μm thick paraffin sections were prepared of formalin-based fixed brain from a non-transgenic wild-type mouse (WT), a transgenic mouse overexpressing human tau with the P301S mutation (P301S) [19] and a tau knock-out mouse (TauKO) [23]. Human fixed tissue from Alzheimer’s disease patient (Braak VI) (Human AD) and non-AD control (Human control) were obtained from the Netherlands Brain Bank, Netherlands Institute for Neuroscience, Amsterdam https://www.brainbank.nl. All material has been collected from donors from whom a written informed consent for brain autopsy and the use of the material and clinical information for research purposes had been obtained by the NBB. After deparaffinization and rehydration of the sections, heat induced antigen retrieval was performed in citrate buffer (pH 6.0) and endogenous peroxidase activity was blocked with 3% hydrogen peroxide. Samples were incubated for 1 h with 0.1μg/ml of primary antibodies mTau2, mTau3, mTau5, mTau8, and mTau9, diluted in antibody diluent with background reducing components (DAKO, S3022). After extensive washing, peroxidase labelled anti-mouse secondary antibody (Envision, DAKO) was applied for 30 min, followed by chromogenic labelling with 3,3-diaminobenzidine (DAB) (DAKO). Slides were counterstained with hematoxylin, dehydrated and permanently mounted (Vectamount, Vector Labs, Burlingame, CA, USA). Sections were imaged with a Nanozoomer slide scanner (Hamamatsu Photonics, Shizuoka, Japan).

Preparation of brain homogenates and sarkosyl extraction

For homogenization, cortical brain regions from different animals were weighed and homogenized in 5× (Cynomolgus, Marmoset macaque, rabbit, mini-pig, beagle, rat), 6× (WT, P301S and TauKO mice) or 10× (human brain control and AD) volume/weight of cold homogenization buffer (10 mM Tris/HCl, 0.8 M NaCl, 10% sucrose, 1 mM EGTA). Human brain was homogenized using a glass/Teflon Potter tissue homogenizer (IKA Works, Inc.) at 1000 rpm. The homogenized material was then centrifuged at 27000× g for 20 min at 4°C. Supernatant, denoted as total homogenate, was collected and kept at –80°C until use. The remaining animal brains were homogenized using FastPrep^®-24 instrument (MP Biomedicals). Next the homogenates were centrifuged at 20000× g for 5 min at 4°C. Supernatant was collected into an ultracentrifuge vial and centrifuged at 34000× g for 20 min at 4°C. Supernatant, denoted as total homogenate, was collected and kept at –80°C until use.

For sarkosyl extraction of tau aggregates from P301S mouse brains, the homogenization procedure was performed as described. However, following ultracentrifugation at 34000× g the supernatant was collected and 1% (weight/volume) of N-laurylsarcosine was added and incubated for 90 min at 37°C with agitation at 900 rpm. After incubation, the homogenate was ultracentrifuged at 180000× g for 1 h at 20°C. Supernatant was collected as “sarkosyl-soluble fraction” and pellet (sarkosyl-insoluble fraction) was resuspended in homogenization buffer. Both were stored at –80°C until use.

Single Molecule Array Assay (Simoa)

Simoa homebrew assays were developed in-house. All reagents used were obtained from Quanterix (Quanterix^®, Billerica, MA). Paramagnetic carboxylated beads were activated in 0.5 mg/ml 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC; Thermo Fisher Scientific) and coated with 0.3 mg/ml of mTau8 according to the manufacturer’s instructions. For detection, PT82 was biotinylated at 60× molar ratio, and used at a final concentration of 1.8μg/ml with a streptavidin-β–D–Galactosidase (SBG) concentration of 200 pM. All samples and calibrators were diluted in detector/sample diluent and run on Simoa HD-1 analyzer as a 2-stepassay.

CSF collection

CSF was collected from WT Bl/6J and P301L mice [24] of 3 and 7 months of age, respectively. CSF collection was performed as described in Liu et al. (2008) [25] with a few modifications. The mice were under isoflurane anesthesia for the whole procedure. Briefly, a glass capillary tube GC150-10 (Harvard Apparatus Limited) connected to a micropump was inserted into the cisterna magna. CSF started flowing into the capillary tube and the pump was set to 50μL/min to speed the procedure. Once the tube was full the pump was stopped, and the tube was removed from the cisterna magna. CSF was collected into a microtube, snap frozen on dry ice, and stored at –80°C until use. Between 8 to 10μL were collected per mouse. Prior to Simoa analysis samples from 2 or 3 mice were pooled together to have enough volume for analysis (minimum of 16μL).

Western blotting

Brain homogenates were diluted in NuPAGEtrademark LDS Sample Buffer and Reducing Agent (Invitrogen) and boiled for 10 min at 75°C with shaking, followed by loading on SDS PAGE (4–12%) (Bio-Rad). After separation, the gel was blotted on a nitrocellulose membrane (Bio-Rad) using Trans-Blot Turbo system (Bio-Rad). The membrane was blocked with TBS-T containing 5% non-fat dry milk and incubated with non-labelled primary antibodies (mTau2, mTau3, mTau5, mTau8, mTau9) overnight at 4°C. Detection was performed with HRP-labelled goat anti-mouse IgG antibody (170-6516, Bio-Rad), incubated for 2 h at RT. SuperSignaltrademark West Dura (Thermo Scientifictrademark) was used to reveal the bands and imaging was done on Amersham Imager 600 (GE Healthcare).

Cell culture and neuronal aggregation assay

Mouse primary hippocampal neurons were isolated from E19 C57Bl/6J (Janvier) and TauKO embryos, dissociated and plated in 96 well plates (μClear, 655946, Greiner) previously coated with poly-L-lysine (P1274, Sigma Aldrich) at 40,000 cells per well. Neurons were kept at 37°C and 5% CO₂ in B-27 and GlutaMax supplemented Neurobasal medium (10888022, Gibco).

The neuronal aggregation assay was performed as described in Soares et al. [26]. In short, neurons were treated with AD-seeds on DIV9 and were kept in culture until DIV23. For immunocytochemistry the cells were fixed with methanol for 15 min at –20°C and blocked with 5% goat serum (Sigma, G9023). Stainings were done with mTau5, mTau8 or mTau9 for visualization of aggregated mouse tau, anti-MAP2 (ab5392, abcam) for neuron identification and anti-GFAP (Z0334, DAKO) for microglia visualization. For Mesoscale analysis of mouse tau aggregates, cells were lysed in RIPA buffer.

RESULTS

Development of new anti-mouse tau antibodies

Antibodies against mouse tau were developed by injecting Balb/c mice with 75μg of the longest isoform of recombinant mouse tau. Based on a screening ELISA, 5 interesting clones were identified—coded mTau2, mTau3, mTau5, mTau8, and mTau9—all of the IgG1κ subtype. hTau62 was raised from mice injected with the longest isoform of recombinant human tau.

Epitope mapping

Epitope mapping was performed on a library of 20-mer peptides with an overlap of 18 aa, covering the complete sequence of human 2N4R tau (Fig. 1). This disclosed an N-terminal epitope for mTau2 ranging from aa7-20: ⁷EFDTMEDHAGDYTL²⁰ and a mid-region epitope for hTau62 ranging from aa159-164: ¹⁵⁹PPGQKG¹⁶⁴. As the epitope mapping was performed on human tau, it is clear that mTau2 does cross-react with human tau, although there are some sequence differences within the epitope between human and mouse tau. This suggests that amino acids 9, 10, 17, and 19 and 20 are not crucial for epitope binding but might be important for binding strength. Figure 3 shows mTau2 binding to full-length mouse and human tau in an indirect ELISA setup and no clear differences in affinity for both tau species are observed. hTau62 is shown to be human specific and does not cross-react with mouse (data not shown) meaning that the amino acids at locations 159 and/or 161 are important for epitope binding and changing the proline and the glycine in the human sequence to respectively a serine and alanine eliminates antibody binding.

Fig. 1

Epitope mapping of mTau2 and hTau62 on human Tau sequence (2N4R). The reactivity of both antibodies to different human tau derived peptides is indicated by the signal strength in the ELISA screening. From this, the epitopes for each antibody can be defined as amino acid sequence ⁷EFDTMEDHAGDYTL²⁰ for mTau2 and amino acid sequence ¹⁵⁹PPGQKG¹⁶⁴ for hTau62. As further described, mTau2 reacts also with mouse Tau and is hence not human nor mouse tau specific, although the amino acid sequence in the epitope region differs between mouse and human. From this we can deduct that amino acids 9, 10, 17, 19, and 20 are less important for antibody binding. hTau62 does not cross-react with mouse tau (data not shown), indicating the importance of amino acids 159 and 161 in antibody binding. Red color: indicates absent amino acids or changes in electrical charge or hydrophobicity of the side chain. Blue color: indicates changes in amino acids without changes in electrical charge or hydrophobicity of the side chain. Yellow color: indicates epitope region of the respective antibodies.

A focused peptide array comprising the region 105–152 in human 2N4R tau and the corresponding region in mouse tau, was used to determine the exact epitopes of mTau3, mTau5, mTau8 and mTau9, and to identify potential human cross-reactivity (Fig. 2). All antibodies bound a similar set of peptides with pronounced signal intensities on the mouse tau peptides. The core epitope sequence ranges from amino acid 133 to 142: ¹³³DRTGNDEKKA¹⁴² and further extension of the fragment at either terminus increases binding. In these stringent conditions, some detectable binding can be observed only on a small subset of longer human peptides. ELISA data (Fig. 3) shows that indeed, at very high antibody concentrations, mTau3 and mTau5 can show cross-reactivity with human tau, albeit that around 60 to 200× higher antibody concentration is needed. This was not observed for antibodies mTau8 and mTau9, which also cross-reacted with a few longer human peptides but did not react with human tau on ELISA, even at high concentrations. There are 3 amino acids within the epitope that differ between the mouse and the human sequence, i.e., amino acids at position 134, 137, and 139 which are respectively arginine, asparagine, and glutamic acid in mouse and glycine, serine, and aspartic acid in human. These differences seem to be enough for mTau8 and mTau9 not to bind human tau, pointing to the importance that those amino acids may have in epitope binding of these antibodies. How each amino acid participates in antibody binding of the different antibodies can only be determined via crystallography, and their importance may be different for each antibody. This can explain why mTau3 and mTau5 do show some extent of human tau binding at a concentration at which mTau8 and mTau9 do not bind, even though they all share the same core epitope sequence.

Fig. 2

Linear intensity profiles of mTau3, mTau5, mTau8, mTau9, and PT79 on human and mouse tau peptides. All four mTau antibodies were tested under high stringency conditions on different peptides derived from mouse and human Tau ranging from amino acid 105–152. Signal intensity measured for the different peptides is represented in different graphs for the human and mouse tau sequence. All antibodies bound a similar set of peptides with pronounced signal intensities on mouse tau peptides. The base epitope sequence was defined as ¹³³DRTGNDEKKA¹⁴² and further extension of this peptide sequence at either terminus increases binding. Some detectable binding can be observed on a small subset of human tau peptides, although a longer epitope sequence is needed. Cross-reactivity seen with mTau8 and mTau9 is neglectable (see also Fig. 3). PT79 did not bind peptides from mouse tau but strong binding was observed to the human tau peptides. The minimal epitope region is defined as ¹³³DGTGSDDK¹⁴⁰ and addition of N-terminal residues ¹³⁰KSKDGTGSDDKK¹⁴¹ increases binding to saturation point. Amino acid sequence on x-axes, with 1 being amino acid 105 and 48 being amino acid 152. Bars are drawn from the starting position to the end position in the peptide at the height of the signal. Base epitope sequence is underlined in mouse tau. Red color: indicates absent amino acids or changes in electrical charge or hydrophobicity of the side chain. Purple color: indicates changes in amino acids without changes in electrical charge or hydrophobicity of the side chain.

Fig. 3

Reactivity of mTau2, mTau3, mTau5, mTau8 and mTau9 with full-length mouse and human Tau in ELISA. Binding of all mTau antibodies to full-length (2N4R) human and mouse tau was assessed in ELISA. Human or mouse full-length tau were coated at 10 ng/ml. Serial dilutions of the respective antibodies was used as primary antibody for tau binding. HT-7 is used as positive control antibody for human tau but negative control antibody for mouse tau reactivity. All antibodies, besides mTau2, bind with equal affinity to mouse Tau in antibody concentrations ranging from 6μg/ml to 60 pg/ml. mTau2 reacts with both mouse and human tau without any obvious difference on affinity. mTau3 and mTau5 show cross-reactivity with human tau at the highest antibody concentrations (from 6μg/ml to 0.6μg/ml). On the contrary mTau8 and mTau9 are specific for mouse tau and show no cross-reactivity with human tau.

PT79 [18] on the other hand showed strong binding to the human tau peptides of this region, without any binding to mouse tau. The minimal epitope region for this antibody is defined at amino acid 133 to 140: ¹³³DGTGSDDK¹⁴⁰, and addition of N-terminal residues ¹³⁰KSKDGTGSDDKK¹⁴¹ increases binding to saturation point.

Profiling via immunohistochemistry

Immunohistochemistry was performed with mTau2, mTau3, mTau5, mTau8, and mTau9 on brain slices from WT, P301S, and TauKO mice, and on human control and AD brain. Using immunohistochemistry, mTau2 (Fig. 4) also showed some degree of human cross-reactivity; tau labelling, however, was only observed on human AD brain and not on human control brain, indicating that the avidity component plays a large role on tau aggregates resulting in mTau2 to stain tau neurofibrillary tangles and neuropil. Furthermore, mTau2 stains all mouse brains. Staining seen in P301S mouse brain is probably a combination of monomeric mouse tau binding and binding to human and mouse tau in aggregates, with avidity playing a role for the human tau binding. Important to note is that the TauKO model used has an EGFP cassette inserted in exon1 of the Mapt gene [23], resulting in the expression of a 31 kDa fusion protein that includes the first 31 amino acids of tau, hence including the epitope region of mTau2 (⁷EFDTMEDHAGDYTL²⁰). This explains the staining profile observed in the TauKO mouse brain, which is different at a cellular level than the staining of WT and P301S mouse brain.

Fig. 4

Reactivity of mTau2 with mouse and human brain tissue in immunohistochemistry. mTau2 binds mouse tau in all mouse brain analyzed. Also, some reactivity with tau aggregates is shown in the P301S mouse brain section. This can be reactivity with mouse tau incorporated in these aggregates or some low avidity for human tau. Note that also a reactivity with Tau–/–brain is observed. The mouse strain used has an EGFP cassette inserted in exon1 of the MAPT gene, resulting in the expression of a 31kDa fusion protein that includes the first 31 amino acids with which mTau2 will react. mTau2 does not react with human tau in healthy control brain, but some reactivity is seen with aggregated tau in human AD brain, underlining the avidity component when for human tau. Scale bars: human healthy control 5 mm, human AD brain and mouse brains 2.5 mm, zoomed pictures 50μm/250μm.

All other mouse tau antibodies (mTau3, mTau5, mTau8, and mTau9) share the same immunoreactivity profile on mouse and human brain sections. Staining was only seen in WT and P301S brain (Fig. 5). As no staining was observed in human (control and AD brain) (Fig. 6), the staining observed in P301S brain is solemnly attributed to endogenous mouse tau, meaning that the neurofibrillary tangles observed do contain a certain level of endogenous mouse tau.

Fig. 5

Reactivity of mTau3, mTau5, mTau8, and mTau9 with mouse brain tissue in immunohistochemistry. All four mouse-specific antibodies react with mouse tau in wild-type mouse brain and P301S brain. No reactivity is found with brain of Tau–/–mice, confirming the specificity of the antibodies for mouse tau. Aggregates in P301S brain are also visualized with these antibodies. Since the antibodies do not show avidity for human tau aggregates in human brain sections of AD patients (Fig. 6), the staining observed in the P301S mouse brain is likely a result of binding to mouse tau in those aggregates. Scale bars: mouse brains 2.5 mm, zoomed in pictures 50μm.

Fig. 6

Reactivity of mTau3, mTau5, mTau8, and mTau9 with human brain tissue in immunohistochemistry. All four mouse-specific antibodies do not have a cross-reactivity with human tau on human brain sections from healthy individuals and AD brain. Scale bars: 5 mm; human AD brain stained with mTau3, mTau5, mTau9 2.5 mm.

Species cross-reactivity

To further explore binding characteristics of the mTau antibodies, each antibody was tested for cross-reactivity with tau from a broader panel of other species. This characterization was performed with a sandwich MSD assay and western blot. The species evaluated were mouse (WT, P301S, and TauKO), rat, rabbit, mini-pig, beagle dog, marmoset monkey, and cynomolgus macaque. Although sequence similarity of tau from these different animal species is between 85 and 98%, small differences in amino acid sequences can result in different binding of the mTau antibodies.

The four mouse-specific antibodies, mTau3, mTau5, mTau8, and mTau9, only reacted with mouse or rat tau, as they share the same amino acid sequence in this region of tau (Fig. 7). The levels of tau measured by these antibodies in the WT and the P301S mouse model, confirm again that only mouse tau is captured in the P301S brain homogenate and that the human tau, albeit overexpressed, does not contribute to the signal observed in MSD. All animals have, similar to human, a glycine at position 134 instead of an arginine, and aspartic acid instead of glutamic acid at position 139. Position 137 varies a lot between species, although all amino acids found at this position are uncharged. The western blot data (Fig. 8) confirm the reactivity with only rat and mouse tau homogenates. Note that antibody heavy and light chains of endogenous IgG are visual in homogenates of TauKO, WT and P301S as a Goat-anti-mouse-IgG secondary antibody is used.

Fig. 7

Reactivity of mTau2, mTau3, mTau5, mTau8, and mTau9 with tau protein from different species. All five mTau antibodies were coated at 1μg/ml to capture tau in total homogenate of dissected brain samples from different animals. Detection was done with PT82 antibody, an antibody that is known to bind to all species with equal affinity as its epitope is conserved. mTau3, mTau5, mTau8, and mTau9 only detect tau in mouse and rat brain homogenates. The levels detected are the same in wild-type non-transgenic animals and P301S animals, indicating that the human protein in the P301S mice is not captured. Tau levels captured with mTau2 in P301S mice are higher than wild-type mice, indicating human tau is also captured, which is in line with previous shown data. mTau2 cross-reacts with homogenates of all species except rabbit, and the reactivity with mini-pig and beagle is only marginal. Dotted line is LOQ of the assay.

Fig. 8

Reactivity of mTau2, mTau3, mTau5, mTau8, and mTau9 with tau protein from different species in western blot. mTau2 reacts with tau from all species tested. In TauKO homogenate mTau2 also reacts with the 31 kDa fusion protein composed of the first 31 amino acids of tau fused to EGFP. mTau3, mTau5, mTau8, and mTau9 are specific for tau from mouse and rat. Lower 28 kDa bands observed in homogenates of TauKO, WT, and P301S mice, and upper 45 kDa band seen on TauKO homogenate correspond to antibody heavy and light chains of endogenous IgG, respectively, as a Goat-anti-mouse-IgG secondary antibody was used.

In contrast to the above-mentioned antibodies, mTau2 reacts in MSD with homogenates of all species but rabbit and the reactivity with mini-pig and dog is only marginal. The reactivity with P301S is higher than WT, meaning that human tau contributes to the signal in P301S mice, confirming once more human cross-reactivity for this antibody. Reactivity with rat is equal to WT, meaning that the two amino acids different between two species (aa14 and aa20) are not essential for antibody binding. On the contrary, in western blot, mTau2 reacts with total homogenate of all species, even the short tau fragment in TauKO mice. It might be that, under denaturing conditions, the amino acid sequence difference from rabbit, mini-pig, and dog, plays less of a role for antibody biding. Note that the epitope region of PT82, which is used as detection antibody, is conserved among all species and cannot cause the differences observed between MSD and western blotting. Moreover, using another detection antibody in the c-terminal region of tau results in similar MSD profiles (data not shown). It is however clear that mTau2 is not mouse tau-specific and can bind tau from multiple mammals under certain conditions.

Application of mouse and human tau-specific antibodies in immunoassay development

Due to the specific nature of mTau8 and hTau62, as being respectively mouse and human tau-specific, these antibodies can be used to distinguish between mouse tau and human tau in mouse models or cellular culture systems where both species arepresent.

When using the same antibody as capture and detection in a sandwich ELISA set up, assays can be created that measure specifically aggregated tau species, as only those species containing multiple tau molecules with the same epitope will be identified. This way, an aggregated mouse tau-specific assay using mTau8 and an aggregated human tau-specific assay using hTau62 were developed. As an example, results are shown of the characterization of total brain homogenate, sarkosyl-soluble and insoluble fractions from young (4 weeks) -before major aggregate deposition- and old (24 weeks) -with strong aggregate deposition- P301S mice (Fig. 9). These mice overexpress 0N4R human tau with the P301S mutation found in FTDP-17, while still expressing endogenous mouse tau [27]. All fractions were analyzed with both aggregation assays. In young mice, human tau-containing aggregates were detected, although at very low levels. Mouse tau-containing aggregates could not be detected in any of the fractions of these mice. In old mice, mouse tau-containing aggregates were detected in both the total brain homogenate and the sarkosyl-insoluble fraction. However, the intensity of the signal was 100-fold lower than with the human aggregation assay. Due to the specificity of these assays, it is clear that mouse tau is part of the aggregates found in old P301S mice. The presence of endogenous mouse tau in tau aggregates of mice that overexpress human forms of tau has been previously described in 3xTg-AD and rTg4510 mouse models [28, 29], thus it is with no surprise that we also observe endogenous tau in the aggregates of the P301S model. Additionally, in old P301S mice, aggregated human tau is also detected in the sarkosyl-soluble fraction, although at much lower levels than found in the insoluble fraction. These are very likely small multimers or oligomers that due to their small size are not pelleting during ultracentrifugation. Similar findings have also been reported on young rTg4510 mice, where 140 kDa tau with some degree of phosphorylation was detected in the sarkosyl-soluble fraction of brain homogenates [30] as detected by AT8 antibody, which recognizes tau protein phosphorylated at both serine 202 and threonine 205 [31, 32], as well as in TBS supernatant fractions [33].

Fig. 9

Example of the use of mouse-specific aggregate assay. Using mTau8 and hTau62, aggregated tau-specific assays for respectively mouse and human tau were made. They can be used to measure the amount of aggregates that contain either of the species or a combination of mouse and human tau. As shown here, tau aggregates from P301S mice are composed of both mouse and human tau. How they relate in abundance is less clear as affinity difference of the antibodies also plays a role. Nevertheless, we can assume that the percentage of mouse tau in the aggregates is lower than the human tau, partially based on the fact that there is a large overexpression of human tau in this model.

Additionally, mTau8 was used to create ultra-sensitive assays that specifically measure mouse tau. For this, the sandwich MSD assay using mTau8 as coating antibody and PT82 as detection antibody (mTau8/PT82), as used in Fig. 7, was transferred to the Simoa platform. The Simoa technology is able to measure proteins at the femtoliter level [34, 35] and has been a valuable tool to measure tau in biological fluids such as CSF, blood, and plasma of patients with different tauopathies [36 –38]. Transfer of assay mTau8/PT82 to Simoa resulted in a 100-fold increase in sensitivity going from a limit of detection of 1000 pg/ml in MSD to 10 pg/ml in Simoa (Fig. 10).

Fig. 10

Development of a mouse-specific Total Tau assay –mTau8/PT82 with high sensitivity. Comparison of the sensitivity of assay mTau8/PT82 on MSD or Simoa. Dotted lines represent LOQ for each platform. There is a 100-fold improvement in sensitivity when the assay is performed on the Simoa platform, going from a limit of detection of 1000 pg/ml in MSD to 10 pg/ml in Simoa.

This level of sensitivity enabled the detection of mouse tau in CSF collected from P301L and WT mice (BL/6J) (Fig. 11). CSF was collected from 15 BL/6J mice and 13 P301L mice and analyzed on Simoa with mTau8/PT82 assay. Because of the low volume of individual samples, pools of CSF from 2 or 3 mice were used and analyzed in the Simoa assay at an 8-fold dilution. As expected, endogenous tau could be measured in both strains at similar levels. This further demonstrates the mouse-specificity of our assay, as the presence of overexpressed mutant human tau did not interfere with the endogenous tau measurements in the P301L model. Even though this was a pilot proof of concept experiment, it clearly demonstrates the potential of our antibodies to measure mouse tau at very low concentrations when applied on a sensitive platform such as Simoa.

Fig. 11

Detection of mouse tau in the CSF of WT mice and P301S (line PS19) mice. CSF was collected from 15 BL/6J mice and 13 P301S mice (line PS19) and individual samples from 2 or 3 mice were pooled together to have enough volume for Simoa analysis. Endogenous tau levels were similar between both strains (p = 0.3124 as determined by Two-tailed Mann Whitney test).

mTau antibodies can specifically detect endogenous mouse tau aggregation in primary neurons

To further demonstrate the potential of our mouse tau-specific antibodies, mTau5, mTau8, and mTau9 were used for immunocytochemistry (ICC) staining of a neuronal tau aggregation model. In this model, tau aggregation is induced in non-transgenic primary mouse neurons by addition of AD-tau seeds extracted from patient brains [26]. To be able to distinguish the newly aggregated mouse tau from the human tau seed it is crucial to have mouse tau-selective antibodies.

All 3 mTau antibodies were able to detect aggregated mouse tau in the non-transgenic neurons in ICC (Fig. 12). In the TauKO neurons the AD-seeds do not promote aggregation as there is no endogenous tau protein present. The lack of signal in these neurons confirms that our antibodies do not recognize the seed and thus the signal observed in the non-transgenic cultures is solely from aggregated mouse tau (Fig. 12).

Fig. 12

Detection of aggregated mouse tau in non-transgenic primary neurons. Primary neurons from non-transgenic (WT) and TauKO mice were treated with 125 ng of AD-seeds on DIV9. On DIV23 the cells were methanol fixed and stained with mouse tau-specific antibodies mTau5, mTau8, and mTau9 (green) to visualize mouse tau aggregates. MAP2 (red) was used as a neuronal marker and GFAP (yellow) was used as a microglial marker. All 3 in-house antibodies detect aggregated mouse tau in WT neurons, without recognizing the AD-seed of human origin, as it can be observed by the lack of signal in the TauKO neurons. Images representative of 3 independent experiments.

In a separate experiment an increasing amount of AD-seeds was added to non-transgenic and TauKO primary neurons and mouse tau aggregation was measured with the developed MSD mTau8/mTau8 sandwich assay. In the non-transgenic neurons mTau8/mTau8 detects mouse tau aggregates, with signal increasing with the amount of AD-seeds used to induce aggregation (Fig. 13). Just as observed on ICC, no signal is observed when AD-seeds are added to TauKO neurons, even at the higher concentrations. Once again, we demonstrate the specificity of antibodies mTau5, mTau8, and mTau9 for mouse tau even in the presence of high amounts of AD-derived tau.

Fig. 13

MSD assay mTau8/mTau8 is specific for mouse tau aggregation. Primary neurons from non-transgenic (WT) and TauKO mice were treated with increasing amounts of AD-seeds on DIV9. On DIV23 the cells were lysed, and the RIPA lysates were analyzed with MSD assay mTau8/mTau8. AD-seeds induced mouse tau aggregation in a concentration-dependent manner, as detected by the mTau8/mTau8 sandwich assay signal. This assay did not pick up any signal in the TauKO neurons, even when 1μg of AD-seeds was added to these cells, demonstrating that all signal observed in the WT cells is solely due to aggregation of endogenous tau.

DISCUSSION

Here we report the development and characterization of different species-specific antibodies—human and mouse—which were developed in mice by injecting the longest recombinant isoform from either mouse or human tau. Four of the reported antibodies, mTau3, mTau5, mTau8, and mTau9 have a high degree of specificity for mouse tau, with mTau8 and mTau9 showing no cross-reactivity with human tau in any of the evaluated assays. Moreover, reactivity with tau from other species, except rat, could not be identified. hTau62 on the other hand proved to be human specific. mTau2, albeit an N-terminal antibody from a region with large sequence difference between different species, is not mouse-specific and shows a large degree of cross-reactivity with human tau.

Studies have shown that the ablation of endogenous mouse tau can accelerate the pathology caused by mutated human tau [28 , 40] and that endogenous mouse tau can be found in tau aggregates from transgenic mouse models [29, 41]. With the developed antibodies, we were able to specifically detect mouse tau-containing aggregates from the brain of 24-week-old P301S mice, showing that endogenous tau is recruited to the newly formed aggregates of P301S human tau. Additionally, we were also able to detect newly formed endogenous tau aggregates from primary mouse neurons after AD-tau seed exposure, without interference from the seed itself. Thus, demonstrating how useful the developed antibodies, especially mTau8, can be used to study how endogenous tau is recruited to aggregates in different model systems as well as distinguish newly formed mouse tau aggregates when human-derived tau seeds are used to induce aggregation.

Using mTau8 we also developed a sandwich immunoassay to detect total tau levels. When applied on a highly sensitive platform such as the Simoa, this assay could detect the presence of mouse tau in the CSF of both WT and P301S (line PS19) mice. Most importantly, tau levels detected with this assay were similar between the two mice, highlighting once again the mouse tau-specificity of our antibody.

In conclusion, we have shown the development of mouse tau-specific antibodies and their application on different qualitative and quantitative assays that can be used to detect mouse tau monomers and mouse tau aggregates specifically without the interference of human tau in the read out. Moreover, our antibodies can potentially be used together with phosphorylation-specific antibodies to measure phosphorylated species of mouse tau.

These can be very important tools to better interpret the results obtained from different model systems as well as to study the role of endogenous tau in tau aggregation and pathology observed in the diverse mouse models available. It can thereby contribute to better characterization and selection of drug candidates that target tau pathology.

Footnotes

ACKNOWLEDGMENTS

The authors have no acknowledgments to report.

FUNDING

The authors have no funding to report.

CONFLICT OF INTEREST

All authors are working for Janssen Research and Development, a pharmaceutical company of Johnson & Johnson.

Some authors are J&J equity holders as J&J employees.

DATA AVAILABILITY

The data supporting the findings of this study are available on request from the corresponding author.

References

Binder

, Frankfurter

, Rebhun

(1985) The distribution of tau in the mammalian central nervous system. J Cell Biol 101, 1371–1378.

Goedert

, Spillantini

, Jakes

, Rutherford

, Crowther

(1989) Multiple isoforms of human microtubule-associated protein tau: Sequences and localization in neurofibrillary tangles of Alzheimer’s disease. Neuron 3, 519–526.

Drubin

, Kirschner

(1986) Tau protein function in living cells. J Cell Biol 103, 2739–2746.

Tashiro

, Hasegawa

, Ihara

, Iwatsubo

(1997) Somatodendritic localization of phosphorylated tau in neonatal and adult rat cerebral cortex. Neuroreport 8, 2797–2801.

Sultan

, Nesslany

, Violet

, Bégard

, Loyens

, Talahari

, Mansuroglu

, Marzin

, Sergeant

, Humez

, Colin

, Bonnefoy

, Buée

, Galas

M-C

(2011) Nuclear tau, a key player in neuronal DNA protection. J Biol Chem 286, 4566–4575.

Pallas-Bazarra

, Jurado-Arjona

, Navarrete

, Esteban

, Hernández

, Ávila

, Llorens-Martín

(2016) Novel function of tau in regulating the effects of external stimuli on adult hippocampal neurogenesis. EMBO J 35, 1417–1436.

Guo

, Noble

, Hanger

(2017) Roles of tau protein in health and disease. Acta Neuropathol 133, 665–704.

Arendt

, Stieler

, Holzer

(2016) Tau and tauopathies. Brain Res Bull 126, 238–292.

Feany

, Dickson

(1996) Neurodegenerative disorders with extensive tau pathology: A comparative study and review. Ann Neurol 40, 139–148.

10.

Wang

, Mandelkow

(2015) Tau in physiology and pathology. Nat Rev Neurosci 17, 22–35.

11.

Nelson

, Alafuzoff

, Bigio

, Bouras

, Braak

, Cairns

, Castellani

, Crain

, Davies

, del Tredici

, Duyckaerts

, Frosch

, Haroutunian

, Hof

, Hulette

, Hyman

, Iwatsubo

, Jellinger

, Jicha

, Kövari

, Kukull

, Leverenz

, Love

, Mackenzie

, Mann

, Masliah

, McKee

, Montine

, Morris

, Schneider

, Sonnen

, Thal

, Trojanowski

, Troncoso

, Wisniewski

, Woltjer

, Beach

, Tredici del

, Duyckaerts

, Frosch

, Haroutunian

, Hof

, Hulette

, Hyman

, Iwatsubo

, Jellinger

, Jicha

, Kövari

, Kukull

, Leverenz

, Love

, Mackenzie

, Mann

, Masliah

, McKee

, Montine

, Morris

, Schneider

, Sonnen

, Thal

, Trojanowski

, Troncoso

, Wisniewski

, Woltjer

, Beach

(2012) Correlation of Alzheimer disease neuropathologic changes with cognitive status: A review of the literature. J Neuropathol Exp Neurol 71, 362–381.

12.

Jankowsky

, Zheng

(2017) Practical considerations for choosing a mouse model of Alzheimer’s disease. Mol Neurodegener 12, 89.

13.

Onos

, Sukoff Rizzo

, Howell

, Sasner

(2016) Toward more predictive genetic mouse models of Alzheimer’s disease. Brain Res Bull 122, 1–11.

14.

Poorkaj

, Kas

, D’Souza

, Zhou

, Pham

, Stone

, Olson

, Schellenberg

(2001) A genomic sequence analysis of the mouse and human microtubule-associated protein tau. Mamm Genome 12, 700–712.

15.

Clavaguera

, Akatsu

, Fraser

, Crowther

, Frank

, Hench

, Probst

, Winkler

, Reichwald

, Staufenbiel

, Ghetti

, Goedert

, Tolnay

, Frank

, Fraser

, Ghetti

, Crowther

, Probst

, Hench

, Akatsu

, Staufenbiel

, Goedert

, Tolnay

, Reichwald

(2013) Brain homogenates from human tauopathies induce tau inclusions in mouse brain. Proc Natl Acad Sci U S A 110, 9535–9540.

16.

Peeraer

, Bottelbergs

, Van Kolen

, Stancu

I-CC

, Vasconcelos

, Mahieu

, Duytschaever

, Ver Donck

, Torremans

, Sluydts

, Van Acker

, Kemp

, Mercken

, Brunden

, Trojanowski

, Dewachter

, Lee

VMY

, Moechars

(2015) Intracerebral injection of preformed synthetic tau fibrils initiates widespread tauopathy and neuronal loss in the brains of tau transgenic mice. Neurobiol Dis 73, 83–95.

17.

Guo

, Narasimhan

, Changolkar

, He

, Stieber

, Zhang

, Gathagan

, Iba

, McBride

, Trojanowski

, Lee

VMY

(2016) Unique pathological tau conformers from Alzheimer’s brains transmit tau pathology in nontransgenic mice. J Exp Med 213, 2635–2654.

18.

Vandermeeren

, Borgers

, van Kolen

, Theunis

, Vasconcelos

, Bottelbergs

, Wintmolders

, Daneels

, Willems

, Dockx

, Delbroek

, Marreiro

, ver Donck

, Sousa

, Nanjunda

, Lacy

, van de Casteele

, van Dam

, de Deyn

, Kemp

, Malia

, Mercken

(2018) Anti-tau monoclonal antibodies derived from soluble and filamentous tau show diverse functional properties in vitro and in vivo , J Alzheimers Dis 65, 265–281.

19.

Yoshiyama

, Higuchi

, Zhang

, Huang

, Iwata

, Saido

, Maeda

, Suhara

, Trojanowski

, Lee

VMY

(2007) Synapse loss and microglial activation precede tangles in a P301S tauopathy mouse model. Neuron 53, 337–351.

20.

Köhler

, Milstein

(1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495–497.

21.

Timmerman

, Puijk

, Meloen

(2007) Functional reconstruction and synthetic mimicry of a conformational epitope using CLIPS^TM technology. J Mol Recognit 20, 283–299.

22.

Langedijk

JPM

, Zekveld

, Ruiter

, Corti

, Back

(2011) Helical peptide arrays for lead identification and interaction site mapping. Anal Biochem 417, 149–155.

23.

Tucker

, Meyer

, Barde

(2001) Neurotrophins are required for nerve growth during development. Nat Neurosci 4, 29–37.

24.

Terwel

, Lasrado

, Snauwaert

, Vandeweert

, van Haesendonck

, Borghgraef

, van Leuven

(2005) Changed conformation of mutant tau-P301L underlies the moribund tauopathy, absent in progressive, nonlethal axonopathy of tau-4R/2N transgenic mice. J Biol Chem 280, 3693–3973.

25.

Liu

, Duff

(2008) A technique for serial collection of cerebrospinal fluid from the cisterna magna in mouse. J Vis Exp, 960.

26.

Soares

, Ferreira

, Mariën

, Delay

, Lee

, Trojanowski

, Moechars

, Annaert

, de Muynck

(2021) PIKfyve activity is required for lysosomal trafficking of tau aggregates and tau seeding. J Biol Chem 296, 100636.

27.

Allen

, Ingram

, Takao

, Smith

, Jakes

, Virdee

, Yoshida

, Holzer

, Craxton

, Emson

, Atzori

, Migheli

, Crowther

, Ghetti

, Spillantini

, Goedert

(2002) Abundant tau filaments and nonapoptotic neurodegeneration in transgenic mice expressing human P301S tau protein. J Neurosci 22, 9340–9351.

28.

Andorfer

, Kress

, Espinoza

, De Silva

, Tucker

, Barde

, Duff

, Davies

(2003) Hyperphosphorylation and aggregation of tau in mice expressing normal human tau isoforms. J Neurochem 86, 582–590.

29.

Ren

, Lin

, Sanchez

, Ceballos

, Polydoro

, Spires-Jones

, Hyman

, Dickson

, Sahara

(2014) Endogenous tau aggregates in oligodendrocytes of rTg4510 mice induced by human P301L tau. J Alzheimers Dis 38, 589–600.

30.

Berger

, Roder

, Hanna

, Carlson

, Rangachari

, Yue

, Wszolek

, Ashe

, Knight

, Dickson

, Andorfer

, Rosenberry

, Lewis

, Hutton

, Janus

(2007) Accumulation of pathological tau species and memory loss in a conditional model of tauopathy. J Neurosci 27, 3650–3662.

31.

Mercken

, Vandermeeren

, Lübke

, Six

, Boons

, Van de Voorde

, Martin

, Gheuens

(1992) Monoclonal antibodies with selective specificity for Alzheimer Tau are directed against phosphatase-sensitive epitopes. Acta Neuropathol 84, 265–172.

32.

Malia

, Teplyakov

, Ernst

, Wu

, Lacy

, Liu

, Vandermeeren

, Mercken

, Luo

, Sweet

, Gilliland

(2016) Epitope mapping and structural basis for the recognition of phosphorylated tau by the anti-tau antibody AT8. Proteins 84, 427–434.

33.

Sahara

, de Ture

, Ren

, Ebrahim

, Kang

, Knight

, Volbracht

, Pedersen

, Dickson

, Yen

, Lewis

(2013) Characteristics of TBS-extractable hyperphosphorylated tau species: Aggregation intermediates in rTg4510 mouse brain. J Alzheimers Dis 33, 249–263.

34.

Rissin

, Kan

, Campbell

, Howes

, Fournier

, Song

, Piech

, Patel

, Chang

, Rivnak

, Ferrell

, Randall

, Provuncher

, Walt

, Duffy

(2010) Single-molecule enzyme-linked immunosorbent assay detects serum proteins at subfemtomolar concentrations. Nat Biotechnol 28, 595–599.

35.

Wilson

, Rissin

, Kan

, Fournier

, Piech

, Campbell

, Meyer

, Fishburn

, Cabrera

, Patel

, Frew

, Chen

, Chang

, Ferrell

, von Einem

, McGuigan

, Reinhardt

, Sayer

, Vielsack

, Duffy

(2016) The Simoa HD-1 analyzer: A novel fully automated digital immunoassay analyzer with single-molecule sensitivity and multiplexing. J Lab Autom 21, 533–547.

36.

Kovacs

, Andreasson

, Liman

, Regelsberger

, Lutz

, Danics

, Keller

, Zetterberg

, Blennow

(2017) Plasma and cerebrospinal fluid tau and neurofilament concentrations in rapidly progressive neurological syndromes: A neuropathology-based cohort. , , Eur J Neurol 24, 1326–e77.

37.

Foiani

, Woollacott

, Heller

, Bocchetta

, Heslegrave

, Dick

, Russell

, Marshall

, Mead

, Schott

, Fox

, Warren

, Zetterberg

, Rohrer

(2018) Plasma tau is increased in frontotemporal dementia. J Neurol Neurosurg Psychiatry 89, 804–807.

38.

Fossati

, Ramos Cejudo

, Debure

, Pirraglia

, Sone

, Li

, Chen

, Butler

, Zetterberg

, Blennow

, de Leon

(2019) Plasma tau complements CSF tau and P-tau in the diagnosis of Alzheimer’s disease. Alzheimers Dementia (Amst) 11, 483–492.

39.

Heraud

, Brion

J-P

, Leroy

, Yilmaz

, De Decker

, Kabova

, Suain

, Ando

, Authelet

, Leroy

, Heraud

, Kabova

, Yilmaz

, Authelet

, Suain

, De Decker

, Brion

J-P

(2010) Deletion of murine tau gene increases tau aggregation in a human mutant tau transgenic mouse model. Biochem Soc Trans 38, 1001–1005.

40.

LaFerla

, Le

, Rodriguez-Ortiz

, Estrada-Hernandez

, Medeiros

, Baglietto-Vargas

, Kitazawa

, Green

, Le

, Estrada-Hernandez

, Rodriguez-Ortiz

, Medeiros

, Green

, LaFerla

(2014) Endogenous murine tau promotes neurofibrillary tangles in 3xTg-AD mice without affecting cognition. Neurobiol Dis 62, 407–415.

41.

Sydow

, Mandelkow

E-M

(2010) “Prion-like” propagation of mouse and human tau aggregated in an inducible mouse model of tauopathy. Neurodegener Dis 7, 28–31.