Co-Expression of Three Wild-Type 3R-Tau Isoforms Induces Memory Deficit via Oxidation-Related DNA Damage and Cell Death: A Promising Model for Tauopathies

Abstract

The three isoforms of 3R-tau are predominantly deposited in neurons bearing neurofibrillary tangles in Alzheimer’s disease (AD), while only 3R-tau accumulation has been detected in Pick’s disease (PiD), suggesting the involvement of 3R-tau in neurodegeneration. However, both the role and the molecular mechanism of 3R-tau in neurodegeneration are elusive. Here, we co-expressed three isoforms of human wild-type 3R-tau in adult mouse hippocampal to mimic the pathologic tau accumulating observed in PiD patients. We found that co-expressing three 3R-tau isoforms induced hyperphosphorylation and accumulation of tau proteins; simultaneously, the mice showed remarkable neuron death with synapse and memory deficits. Further in vitro and in vivo studies demonstrated that co-expressing 3R-tau isoforms caused oxidative stress evidenced by an increased malondialdehyde, and the decreased superoxide dismutase and glutathione peroxidase; the 3R-tau accumulation also induced significant glial activation and DNA double-strand breaks (DSBs). Notably, the toxic effects of 3R-tau accumulation were efficiently reversed by administration of antioxidants Vitamin E (VitE) and Vitamin C (VitC), respectively. These data reveal that intracellular accumulation of 3R-tau isoforms in adult brain induces significant neuron death and memory deficits with the mechanism involving oxidation-mediated DSBs; and the antioxidants VitE and VitC can efficiently attenuate the toxicities of 3R-tau. Given that no significant cell death has been detected in the currently available wild-type tau-accumulating models, co-expressing 3R-tau isoforms could be a promising model for drug development of tauopathies, such as PiD.

Keywords

3R-tau DNA double-strand breaks oxidative stress tauopathies

INTRODUCTION

Tau belongs to the family of microtubule-associated proteins and plays an essential role in the establishment of neuronal cell polarity, axonal outgrowth, axon transport, and the maintenance of axonal morphology [1, 2]. In the mid-1980 s, it was found that the abnormally hyperphosphorylated tau is the major protein component of the neurofibrillary tangles in the brains of Alzheimer’s disease (AD), indicating that tau may play a deleterious role in neurodegeneration [3]. Since then, one major group of neurodegenerative proteinopathies has been associated with the deposition of pathological tau protein, hence called “tauopathies” [4]. At present, over twenty tauopathies have been discovered, and these tauopathies are an important cause of cognitive and behavioral impairment in the aging population [4, 5].

The microtubule-associated protein tau (MAPT) is encoded by a single gene located on chromosome 17q21, alternative RNA splicing of exons 2, 3, and 10 can produce six different tau isoforms in the central nervous system [6 –8]. This results in isoforms that have 0, 1, or 2 N-terminal inserts (0 N, 1 N, or 2 N), and either 3 or 4 C-terminal repeated sequences [8]. Tau contains one microtubule binding domain, which consists of 3 or 4 microtubule binding repeats. Therefore, tau is divided into 3R-tau and 4R-tau. Tauopathies are divided into those containing 3R, 4 R, or both species of tau proteins. 4R-tau is predominantly present in corticobasal degeneration and progressive supranuclear palsy, both 3R-tau and 4R-tau are found in AD and frontotemporal dementia, while PiD only accumulates 3R-tau [9]. So far, many transgenic mouse models have been established to study the role of 4R-tau in tauopathies [10, 11]. In addition, the tangle-bearing neurons in the AD brain are predominantly 3R-tau-positive [12, 13], while PiD is mainly caused by 3R-tau accumulation, suggesting the involvement of 3R-tau in neurodegeneration. However, the cause-and-effect relation of 3R-tau with neurodegeneration is still poorly understood.

PiD is a rare tauopathy presenting with dementia [14]. PiD is characterized by cortical atrophy associated with neuronal loss, gliosis, and formation of tau-positive, globular, intra-neuronal inclusions denominated Pick bodies [15]. Pick bodies are composed of 3R-tau, which are immunoreactive with antibodies against p-Tau epitopes, neurofilaments, and ubiquitin, while negative with the Gallyas silver stain and Thioflavin-S [16]. Though mutant 3R-tau found in familial PiD cases could induce neuron death [17], the mechanisms underlying wild-type tau-induced majority sporadic PiD are not clear.

Oxidative stress not only contributes to the normal aging, but also to the pathophysiology of many neurodegenerative disorders, such as PiD [18]. Oxidative stress induced by reactive oxygen species (ROS) and free radical generation causes oxidative damage to the central nervous system, including lipids, protein and DNA injury, energy deficiency, inflammation, and tau hyperphosphorylation [19, 20]. Increasing evidence show that oxidative DNA damage induces cell cycle reentry, transcription abnormality, and cell death [21]. A recent study demonstrated that DNA double-strand breaks (DSBs) were significantly higher in hippocampus and neocortex in human amyloid precursor protein (hAPP) transgenic mice than the wild-type mice, indicating that amyloid-β (Aβ) may be a causal factor of oxidative DNA damage [22]. Notably, tau reduction could prevent Aβ-induced DSBs in neurons [22], suggesting that tau plays an essential role in mediating Aβ-induced DNA damages.

In the present study, we investigated whether and how accumulation of wild-type 3R-tau plays a role in neurodegeneration. We co-expressed three isoforms of human 3R-tau in the adult mouse brain, which does not express endogenous 3R-tau, to mimic the tau expressing/accumulating pattern in PiD. We found that co-expression of three 3R-tau caused memory impairment and synapse deficits in mice. Furthermore, accumulation of wild-type 3R-tau induced glial activation, significant cell loss with remarkably increased DSBs and oxidative stress. Supplement of antioxidants [Vitamin E (VitE) and Vitamin C (VitC)] effectively attenuated 3R-tau-induced oxidative stress, DNA damage, and cell death with improvement of learning and memory capacities.

MATERIALS AND METHODS

Plasmids, viruses, reagents, and antibodies

The plasmids pEGFP-tau-0N3R, pEGFP-tau-1N3R, and pEGFP-tau-2N3R, encoding different isoforms of human 3R-tau, were generous gifts from Fei Liu (Nantong University, China). The plasmids DsRed-tau-1N3R and flag-tau-2N3R and virus AAV-CMV-eGFP-tau-0N3R, AAV-CMV-eGFP-tau-1N3R, AAV-CMV-eGFP-tau-2N3R, AAV-CMV-eGFP-tau-0N4 R, AAV-CMV-eGFP-tau-1N4 R, and AAV-CMV-eGFP-tau-2N4 R were constructed from OBio Biologic Technology Co., Ltd [23]. (+)-α-Tocopherol acid succinate (C29H50O2, Vitamin E; Sigma, T3251) and (+)-L-Ascorbic acid sodium (C6H8O6, Vitamin C; Sigma, 134-03-2) were used in the forms of 99.0% pure powder as antioxidants for cell culture and animal studies. All other reagents were obtained from Sigma-Aldrich. The information for antibodies employed was listed in Supplementary Table 1.

Primary hippocampal neuron culture

For primary neuron cultures, 18 days embryonic (E18) rat hippocampus were softly cut on ice in D-Hank’s buffered saline solution, and then suspended three times in 0.25% trypsin solution at 37°C, 5 min each time. Cells were seeded at a density of 3–4×10⁴ cells per well on 12-well plates coated with Poly-D-Lysine/Laminin (Bioscience) in neurobasal medium (Invitrogen) supplemented with 2% B27 and 0.5 mM glutamine. Half of the culture medium was changed every 3 days with neurobasal medium supplemented with 2% B27 and 0.5 mM glutamine. All reagents for cell culture were purchased from Thermo Fisher Scientific. The primary neurons were grown in a humid atmosphere containing 5% CO₂ at 37°C. More than 90% of the cells were neurons after they were cultured. At 2 to 3 div, neurons were infected with tau virus according to the protocol.

Animals, stereotaxic surgery, and drug treatment

C57BL/6 mice (2 months old, male, 30 ± 5 g) were purchased from the Experimental Animal Central of Wuhan University. Mice were raised in standard cages at a temperature of 25 ± 1°C on a 12 h light/12 h dark cycle with free access to water and food. All animal experiments were performed according to protocols approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology.

Mice were anesthetized with chloral hydrate and placed on a stereotaxic apparatus (Narishige, Japan). After routine disinfection with iodophor, the scalp was incised in the middle. Stereotaxic localization of the bilateral CA3 region (posterior 2.2 mm, lateral 2.7 mm, and ventral 2.3 mm) was performed according to the mouse brain localization map. Drilling the skull with a dental drill, using a microinjection system (World Precision Instruments), the mixed AAV-eGFP-3R-Tau (1 μl, 3.78 × 10¹² viral genome/ml) was injected in the hippocampus CA3 region at a rate of 0.125 μl/min, while the control group was given an equal dose of vector. The needle was retained for 10 min before withdrawal and the skin was sutured.

To investigate the role of antioxidants in improving tau-induced cognitive impairment, the mice were randomly divided into six groups and administered with antioxidant VitC (100 mg/kg) or VitE (10 mg/kg) for 14 consecutive days via intraperitoneal injection. VitC was dissolved in normal saline (NS) while VitE was dissolved in dimethyl sulfoxide (DMSO) and the control was injected with NS or DMSO. For primary neuron culture studies, VitC at 125 μM, 250 μM, and 500 μM, and VitE at 12.5 μM, 25 μM, and 50 μM were applied, respectively.

Immunohistochemistry and immunofluorescence

For immunohistochemical staining, the mice were rapidly perfused with 150 ml 0.9% NS and slowly infused with 300 ml 4% paraformaldehyde (PFA). Then coronal sections (25 μm) were cut by using a cryostat microtome. After 3 times rinses with PBS, cryosections were permeabilized in PBS containing 3% H₂O₂ and 0.5% Triton X-100, following by blocked nonspecific binding with 5% bovine serum albumin (BSA) for 1 h. The free-floating brain sections were incubated with primary antibodies at 4°C overnight, biotin-labeled secondary antibodies and horseradish peroxidase-labeled third antibodies each for 1 h at room temperature, and visualized with the DAB tetrachloride system. The images were taken under the same settings of the respective conditions using a microscope (Olympus BX60, Tokyo, Japan). For immunofluorescent staining, cultured neurons were fixed in 4% PFA for 15 min and permeabilized in 0.5% Triton X-100 in PBS for 15 min, while brain sections were prepared as previously described. The primary antibodies were then incubated at 4°C overnight. The Alexa Fluor 488/568-conjugated secondary antibodies (1 : 1000, Invitrogen) were used for 1 h at room temperature, and then DAPI (1 : 1,000) for 15 min. Immunofluorescence images were observed by using a confocal laser scanning microscope (Carl Zeiss, Jena, Germany) and acquired using ZEN 2009 software (Carl Zeiss, Jena, Germany).

Nissl staining

The 25 μm PFA fixed frozen sections were dried in air for 1 h, then placed onto the gelatin-coated slide and immersed into 1% toluidine blue for 1 min. Remaining dye was washed out by water, followed by decolorization in a concentration of ethanol series (50% for 10 min, 70% for 10 min, 95% for 10 min, 99.9% for 10 min). Finally, sections were cleared in xylene (20 min). The images were photographed using an optical microscope (Olympus BX60, Tokyo, Japan) and the Nissl-stained neurons were counted.

SOD, MDA, and GSH–PX measurement

Superoxide dismutase (SOD; U/mL), malondialdehyde (MDA; nmol/mL), and glutathione peroxidase (GSH-Px; mIU/mL) were measured by using kits following the manufacturer’s instructions (Nanjing Jiancheng Bioengineering Institute, Nanjing, PR China).

Spatial learning and memory tests

The Morris water maze (MWM) and contextual fear conditioning tests were used to test spatial learning and memory. For the MWM, the apparatus consisted of a large circular pool (120 cm diameter and 50 cm height, filled with 30 cm of 22 ± 2°C opaque water). A hidden escape platform (1.2 cm below the surface of the water) was kept constant in the third quadrant during training. All mice were trained to search the platform for 6 consecutive days from 6 : 00 PM to 10 : 00 PM; test trials were performed four trials per day. In each trial, the mice were faced toward the pool wall and released arbitrarily to find the platform in 60 s. Mice were guided to the platform and forced to stay there for 20 s if they could not locate the platform in 60 s. The probe test was conducted 1 d (day 7) after the last place trial to determine the ability of spatial memory retention. The platform was removed and the mice were tested for the latency to find the platform in the target zone, and the numbers of platform crossings were recorded.

The contextual fear conditioning was applied one week after MWM testing. The mice were trained and tested in a square chamber with an electric grid floor. In the training phase (day 1), mice were habituated into the chamber for 3 min, then given a 2 s, 0.5 mA foot shock, and another foot shock was given after 1 min interval, repeated three times. Then the mice were returned to their home cages. In the contextual test (day 2), mice were put back into the same chamber and the percent of freezing time without the stimulus for 3 min was measured.

Nesting behavior test

The nesting ability of mice was correlated with the hippocampus and forebrain cortex function. Each mouse was placed in a cage with enough food and water, and six cottons of the same size (5×5 cm²) were put into each cage. The formation of the nest was examined and the scores were given the next day [24].

Anxiety-like behavior tests

The open field test (OFT) [25] was used to test the locomotor activity and the anxiety behavior of the mice. The mouse was placed in the center of an open field apparatus (60×60×40 cm), and their free-moving behavior was monitored for 5 min and the total distance traveled, the speed and the duration time in the central region were analyzed by using the Noldus video tracking system (Ethovision).

The elevated plus maze test (EPM) [26] was also used for analysis of anxiety-like behaviors. The apparatus was consisted of four arms (two open arms and two closed arms: 35×6×20 cm) with a square area (6×6 cm) in the middle. Mice were individually placed on the platform facing an open arm, and allowed to free explore for 5 min. The numbers of entries into the open arms and the total amount of time spent in open arms were recorded and analyzed by the Noldus video tracking program (Ethovision).

If the same animal was exposed to all above-mentioned behavioral tests, the sequence of the tests was arranged from mild to strong stimulations, i.e., nest building, OFT, EPM, MWM, and fear conditioning, to maximally avoid interference of the previous test to the latter ones.

Electrophysiological recordings

Mice were deeply anesthetized by 6% chloral hydrate, brains were removed, and cut into horizontal sections of 350 μm thickness by a vibratome (VT1000 S, Leica, Nussloch, Germany) in ice-cold artificial cerebrospinal fluid (ACSF) containing (in mM): NaCl 124; KCl 3.0; MgCl₂ 1.0; CaCl₂ 2.0; NaH₂PO₄ 1.25; NaHCO₃ 26; glucose 10; saturated with 95% O₂, 5% CO₂ (4°C, pH 7.4). Then the hippocampal slices were incubated at 32°C for 30 min in ACSF, and allowed to equilibrate to room temperature for > 30 min. After 2 h incubation, individual slices were laid down over an 8×8 microelectrode array in the bottom planar, each 50×50 μm in size, with an interpolar distance of 150-μm (MED-P515A; Alpha MED Sciences, Kadoma, Japan) and kept submerged in ACSF (2 ml/min; 30°C) with a platinum ring glued by a nylon silk. Voltage signals were acquired using the MED64 System (Alpha MED Scientific). The field excitatory postsynaptic potentials (fEPSPs) were recorded from the CA1 neuron in the hippocampus by stimulating Schaeffer fibers from CA3. Stimulation intensity was adjusted to evoke fEPSP amplitudes that were 40% of maximal size. LTP was induced by applying three train of high frequency stimulation (HSF, 100 Hz) [27].

Golgi staining

The animals were sacrificed by overdose of 6% chloral hydrate, and intracardially perfused with 150 ml 0.9% normal saline (NS), followed by 500 ml 4% formaldehyde, and then perfused with 500 ml Golgi fixative for 2 h in the dark. Subsequently, the brain tissues were removed and further incubated in the above dying solution for 3 days and then the tissues were transferred into 1% silver nitrate solution for 3 days in dark. The brain tissues were cut into horizontal sections of 40 μm thicknesses using a Vibratome (VT1000 S; Leica, Germany) and mounted on the gelatin-coated slide. The sections were dehydrated in successive alcohol and cleared in xylene, and then sealed with neutral gum. The Golgi-impregnated neurons and dendritic branches in hippocampal CA3 were imaged by using a bright-field microscopy (Axioplan 2; Zeiss, Brighton, MI). The dendritic spines from at least 60 neurons per group were counted in a blind manner using Neurolucida software (MicroBrightField, Williston, VA) [28].

Western blotting

The dissected hippocampal CA3 tissues or harvested primary hippocampal neurons were lysed with RIPA buffer (Beyotime, China). The protein concentration was measured by the BCA Protein Assay Kit (Pierce, Rockford, IL, USA). The equal amounts of protein were separated by electrophoresis on 10% sodium dodecyl sulfate polyacrylamide gels (SDS-PAGE) and transferred to nitrocellulose (NC) membranes, then blocked for 1 h with 5% nonfat milk at room temperature. After incubating with primary antibodies at 4°C overnight, membranes were incubated with anti-rabbit or anti-mouse IgG conjugated to IRDyetrademark (800CW) (1 : 10000) for 1 h at room temperature. Immunoreactive bands were scanned and visualized with the Odyssey Infrared Imaging System (Li-Cor Biosciences, Lincoln, NE, USA) [29] and quantitatively analyzed by using ImageJ software (NIH).

NADH and ratio of NAD+/NADH measurement

Frex and SoNar sensors were used to measure the intracellular NADH and the ratio of NAD⁺/NADH as indicated previously [30]. Briefly, HEK293 cells were transfected with Frex and SoNar using Neofect (Biotech Co., Ltd. Beijing). After 48 h, the cells were harvested by trypsinization and centrifuged at 250×g for 5 min. The pellets were washed and then resuspended with PBS-Glucose (HyClone) at 37°C. The cell suspensions were added into 96-well plate using an 8-channel electronic pipette and the cell number of each well was about 2.0–10×10⁴. The fluorescence intensity was recorded by using a Multi-Mode Microplate Reader (BioTek) with dual-excitations at 420 nm and 485 nm and single emission at 520 nm. The background fluorescence was corrected by subtracting the intensity of the samples not expressing Frex and SoNar.

Statistical analyses

All data were collected and analyzed in a blinded manner. Statistical analysis of the results was performed by using SPSS 12.0 statistical software (SPSS Inc. Chicago, IL, USA). p-values were calculated with unpaired Student’s t-test for two-group comparisons, one-way or two-way ANOVA. The data were expressed as mean±SEM and p < 0.05 were considered statistically significant.

RESULTS

Co-expressing three 3R-tau isoforms in vitro and in vivo and the identification

The adult mouse does not express 3R-tau in the brain [6 , 31–33], which makes it an ideal model for studying the role of 3R-tau in neurodegeneration. To establish a mouse model that co-expressing three isoforms of human 3R-tau proteins, we infused stereotaxically the mixture of three 3R-tau isoforms (AAV-eGFP-tau-0N3R, AAV-eGFP-tau-1N3R, and AAV-eGFP-tau-2N3R; 1 : 1:1) or the AAV-eGFP vector (same volume and similar titer) into the dorsal hippocampal CA3 of 2-month-old C57BL/6 mice. After 2 months, the expression of 3R-tau isoforms was confirmed by direct immunofluorescence imaging (Fig. 1a) and co-staining of HT-7 (an antibody specifically reacts with human tau) and GFP (Fig. 1c), in which 3R-tau was predominantly deposited in the neuronal soma and the neurites. The expression of three isoforms of 3R-tau was also identified by western blotting using anti-GFP and HT-7 (Fig. 1b).

Fig.1

Simultaneous overexpressing three 3R-tau isoforms can induce accumulation of hyperphosphorylated tau. The mixture of three 3R-tau isoforms (3R; AAV-eGFP-tau-0N3R, or AAV-eGFP-tau-1N3R or AAV-eGFP-tau-2N3R; 1 : 1:1) or the empty vector (Vec) were stereotaxically infused into the mouse hippocampal CA3 (2 months old). a) The representative immunofluorescence imaging of hippocampal CA3 after infusion of AAV-eGFP-hTau for 3 months (DAP1, blue; GFAP, green). b, c) The representative images show robust expression of human tau in dorsal hippocampal CA3 verified by immunofluorescence and western blotting using anti-GFP and HT-7 (specifically probe human tau). d–f) The expressed 3R-tau was hyperphosphorylated at Ser396 and Ser404 measured by western blotting (d) and immunohistochemistry (f). e) The quantitative analysis of (d) (p = 0.315; two-tail t test; at least 3 mice for each group). g) Simultaneous overexpression of 3R-tau isoforms induces intracellular accumulation of tau at AT8 epitope. h) Simultaneous overexpression of 3R-tau isoforms induces two forms of tau inclusion (asterisks: tangle-like inclusions; arrows: accumulation of the dystrophic neurites). At least 3 mice were measured for each group.

To verify the phosphorylation and accumulation statues of the exogenously expressed 3R-tau, we used phosphorylation site-specific antibodies. By western blotting and immunohistochemistry, a significantly increased level of phosphorylated tau at Ser396 and Ser404 was detected in hippocampal CA3 infused with 3R-tau mixture (Fig. 1d–f), and some tangle-like (asterisks) or Pick body-like (arrowheads) structures were detected by using AT8 antibody (Fig. 1g, h). These data together demonstrate that different isoforms of 3R-tau can be efficiently co-expressed both in intro and in vivo, and the co-expression induces intracellular accumulation of the hyperphosphorylated tau.

To confirm that the three isoforms of tau proteins were indeed co-expressed in a same cell, we further conducted in vitro experiment by co-transfection of HEK293T cells or primary hippocampal neurons (7 div) with three isoforms of tau plasmids (0N3R-eGFP, 1N3R-DsRed, and 2N3R-flag). The immunofluorescence images showed that three isoforms of tau proteins were indeed co-localized in the same cells (Supplementary Figure 1a) or the same neurons (Supplementary Figure 1b). These data together confirm that three isoforms human tau can be efficiently co-expressed in the cells or neurons.

Co-expressing 3R-tau isoforms in adult brain induces memory deficit with synapse impairments

To explore the effects of 3R-tau on spatial learning and memory, we employed the MWM test and contextual fear conditioning after co-expressing 3R-tau isoforms in hippocampal CA3 subset for 2 months. By MWM test, we found that expressing 3R-tau significantly increased the latency to find the hidden platform at day 3, 4, 5, and 6 during training (Fig. 2a), suggesting spatial learning deficit. On day 7, the platform was removed to test the spatial memory retention. Compared with the controls, mice expressing 3R-tau showed significantly increased latency to locate the platform target zone (Fig. 2b) with reduced crossings in the target area (Fig. 2c), indicating spatial memory impairments by 3R-tau. By contextual fear conditioning in the separate sets of mice, we did not observe a significant difference between 3R-tau and the empty vector groups during training (Supplementary Figure 2a), but a significantly decreased freezing was shown in 3R-tau-expressing mice during the memory test carried out at the next day after training (Fig. 2d), suggesting an impaired contexture memory by 3R-tau.

Fig.2

Co-expression of three 3R-tau isoforms in adult brain induces learning and memory deficits with synapse dysfunction. A series of behavioral experiments were probed 2 months after transfection of AAV-3R-tau (3R) isoforms or the empty vector (Vec) (n = 12 mice for each group). a–c) The spatial learning and memory was measured by Morris water maze (MWM). During 6 days learning (a), 3R-tau mice showed an increased latency to find the target at day 3 to day 6. During memory test at day 7 by removed platform, 3R-tau mice showed prolonged latency to locate the platform target zone (b) with decreased number of target platform (c) (Two-way ANOVA column factor, F1, 132 = 6.752, p < 0.05; Bonferroni post hoc tests, p = 0.001, latency; p = 0.013, number of target platform crossing). d) Fear conditioning trial was performed one week after the MWM test. 3R-tau mice showed decreased freezing response in memory test (p = 0.009; two-tail t-test). e) 3R-tau mice also showed a significantly decreased nesting behavior (p = 0.006; two-tail t test). After behavioral tests, the mice were sacrificed for the analyses. f–h) 3R-tau inhibited input–output response (f) and the slop of the evoked fEPSP in hippocampal CA3 subset (g, and the suppression was still significant at 60 min after high frequency stimulation (HFS) (h) compared with the vector (p = 0.024; two-tail t-test; n = 8 slices from 3 mice for Vec, and n = 12 slices from 4 mice for 3R). i, j) By Golgi staining, 3R-tau mice showed a significantly decreased spine number in hippocampal CA3 (p = 0.015; two-tail t test; n = 50–60 for each group). In primary hippocampal neurons cultured for 14 div, Co-expression of 3R-tau decreased dendritic numbers (k, l) (p = 0.0012; two-tail t test; n = 35–40 for each group). m, n) Quantitative assessment of the synaptic-associated protein of 3R-tau and vector mice in hippocampal CA3 extracts. Co-expression of 3R-tau isoforms significantly decreased GluA2, GluN1 and GluN2B levels in hippocampal CA3 extracts with no change on the expression of synaptotagmin, synaptophysin, synapsin1, GluN2A, and PSD95 (p = 0.333, synaptotagmin; p = 0.42, synaptophysin; p = 0.5805, synapsin1; p = 0.036, GluA2; p = 0.049, GluN1; p = 0.118, GluN2A; p = 0.0455, GluN2B; p = 0.538, PSD93; p = 0.43, PSD95; two-tail t test; at least 3 mice for each group). o, p) 3R-tau decreased the ratios of p-CREB/CREB and p-CaMK4/CaMK4 without significantly affecting ERK and the ratio of p-ERK/ERK (p = 0.067, CREB; p = 0.0514, p-CREB; p = 0.007, p-CREB/CREB; p = 0.443, CaMKIV; p = 0.046, p-CaMKIV; p = 0.03, p-CaMK4/CaMKIV; p = 0.743, ERK; p = 0.8116, p-ERK; p = 0.243, p-ERK/ERK; two-tail t test; at least 3 mice for each group). Data were expressed as mean±SEM. *p < 0.05 versus Vec.

We also tested nest building, an instinctive behavior to jointly assess general social behavior, cognitive performance, and motor capabilities of the mice [34]. We observed that the score in 3R-tau-expressing mice was much lower than the empty vector controls (Fig. 2e), suggesting an impaired nestled task. No difference in motor ability or anxiety was shown between the 3R-tau and the empty vector control mice in the open field test (Supplementary Figure 2b, c) and the elevated plus maze (Supplementary Figure 2d, e). These data together indicate that expressing 3R-tau isoforms in hippocampal CA3 significantly impairs spatial memory and instinctive functions without changing the emotion status and motor ability of the mice.

To explore the molecular mechanisms underlying the cognitive deficits, we measured the memory-related synaptic plasticity in 3R-tau mice. By ex vivo brain slice electrophysiological recording, we found that 3R-tau accumulation suppressed basal synaptic transmission evidenced by a reduced input–output (IO) curve and an attenuated slope of fEPSP after high-frequency stimulation (Fig. 2f–h). Co-expression of 3R-tau isoforms also significantly reduced spine density in vivo measured by Golgi staining (Fig. 2i, j) and dendrite numbers in primary hippocampal neurons cultured for 14 div (Fig. 2k, l).

To explore the molecular mechanism underlying 3R-tau-induced synapse dysfunction, we analyzed synaptic-related proteins using western blot. Compared with the controls, 3R-tau-expressing mice showed a significantly decrease level of GluA2, GluN1, and GluN2B in extracts of hippocampal CA3 subset (Fig. 2m, n). To further explore the upstream contributor for the reduced synaptic proteins, we measured CREB, a crucial functional protein for memory formation and consolidation [35]. We found that expression of 3R-tau isoforms significantly decreased the ratio of pS133CREB/CREB (Fig. 4o, p). We also measured CaMKIV and ERK, which can phosphorylate CREB. The results showed that pCaMKIV and the ratio of pCaMKIV/CaMKIV decreased, whereas the activity of ERK was not changed in 3R-tau mice (Fig. 2o, p). These data together indicate that neuronal accumulation of 3R-tau induces memory deficits and synapse impairment of mechanisms for involving CREB inactivation.

Co-expressing 3R-tau induces glial cells activation

Studies have shown that the gliosis and reactive microglia can be found in many tauopathies [36 –40]. To explore whether co-expressing wild-type 3R-tau isoforms affects glial cells, immunohistochemical and immunofluorescence staining were performed. The results showed that co-expression of 3R-tau isoforms in hippocampal CA3 subset remarkably increased numbers of GFAP+-astrocytes and Iba1+-microglia compared with the empty vector controls (Fig. 3a–f). Both astrocytes and microglia displayed enlarged or phagocytic phenotypes and they were frequently presented in the proximity of tau-accumulated neurons (Fig. 3e, f). Interestingly, both microglia and astrocytes were co-stained with human tau (Fig. 3g and Supplementary Figure 3a), and the activated microglia cells were frequently encased on 3R-tau-accumulated neurons. These observations indicate that neuronal co-expression of 3R-tau may elicit the toxic effects by activating glial cells.

Fig.3

Co-expressing of three 3R-tau isoforms can activate glial cells. a–d) co-expression of 3R-tau isoforms significantly activated astrocyte and microglia displayed by enlarged or phagocytic phenotypes in the proximity measured by immunohistochemistry and quantitative analyses (p = 0.045, GFAP; p = 0.0095, IBA1; two-tail t test; at least 3 different slices from 3 mice for each group). e, f) Co-immunofluorescence staining of GFP (green) with GFAP (red) or IBA-1 (red) in dorsal hippocampal CA3 (arrow shows microglial encasement to the tau-expressing neuron). g) Co-immunofluorescence staining of IBA-1 (red) with human tau antibody HT7 (cy5) shows microglial encasement to the tau-expressing neuron. Data were expressed as mean±SEM. *p < 0.05, **p < 0.01 versus Vec.

Co-expressing 3R-tau induces remarkable cell loss with oxidative stress and DNA damages

We then studied the effects of 3R-tau isoforms on cell viability. By Nissl staining, we observed that overexpression of 3R-tau isoforms in hippocampal CA3 induced a remarkable neuron loss (Fig. 4a, b and Supplementary Figure 4a). Furthermore, 3R-tau induced a significantly reduced DAPI with increased immunoreactivity of the cleaved caspase-3 in hippocampal CA3 subset (Fig. 4c, d). These data indicate that accumulation of 3R-tau induces cell apoptosis.

Fig.4

Co-expression of 3R-tau isoforms induces remarkable cell loss with oxidative stress and neuronal DSBs. a, b) 3R-tau decreased neuron numbers in hippocampal CA3 but not in CA1 or DG subset measured by Nissl staining (p = 0.211, CA1; p = 0.0017, CA3; p = 0.45, DG; two-tail t test; at least 3 different slices from 3 mice for each group). c, d) 3R-tau significantly increased cleaved caspase-3 level measured by immunofluorescence and optical density analysis (p = 0.009; two-tail t test; at least 3 different slices from 3 mice for each group). e–g) Simultaneous overexpression of 3R-tau isoforms increased MDA (e) (p = 0.048; two-tail t test) with inhibition of SOD (f) (p = 0.0038; two-tail t test) and GSH-Px activity (g) (p = 0.013; two-tail t test; n = 6–8 for each group) compared with the empty vector. h) Hippocampal neurons cultured for 7 or 14 div were infected with AAV-eGFP-3R-tau (3R) or the empty vector (Vec), and an increased nuclear staining of γH2A.x (red) was observed in 3R-tau-expressing neurons (green). i, j) Co-expression of 3R-tau isoforms increased γH2A.x level in hippocampal CA3 measured by co-immunofluorescence imaging and quantitative analysis (p = 0.035; two-tail t test; at least 20 different cells of three individual slices for each group.) k, l) 3R-tau increased 53BP1 in hippocampal CA3 (p = 0.02; two-tail t test; at least 3 mice were analyzed for each group). m, n) The increased expression of 53BP1 protein in 3R-tau mice was confirmed by western blotting (p = 0.041; two-tail t test; at least 3 mice were analyzed for each group). Data were expressed as mean±SEM. *p < 0.05 versus Vec.

Oxidative stress is not only a key contributor to the onset and deterioration of AD [41, 42], but also involved in the pathogenesis of PiD [18], indicating that oxidative stress may play a critical role in tau pathology. To explore the mechanisms underlying the toxic effects of 3R-tau, we measured MDA level and SOD and GSH-Px activity, the markers of ROS. The level of MDA was significantly increased (Fig. 4e), and the activities of SOD and GSH-Px decreased in hippocampal CA3 subset of 3R-tau mice (Fig. 4f, g). These data suggest that accumulation of 3R-tau induces oxidative stress.

Peroxidation is one of the factors causing DNA damage, and DSBs are one of the most lethal forms of DNA damage [43 –45]. To further verify the toxic effects of 3R-tau accumulation, we measured γH2A.X, a well-recognized marker for DSBs, both in vitro and in vivo. We found that co-expressing 3R-tau significantly increased γH2A.X foci and the γH2A.X-positive cells in hippocampal neurons cultured for 7 div or 14 div (Fig. 4h). In mouse hippocampal, co-expressing 3R-tau remarkably increased γH2A.X-immunoreactive focus measured by immunofluorescence staining (Fig. 4i, j). Co-expressing 3R-tau isoforms also significantly increased 53BP1 (Fig. 4k–n), another marker of DSBs [46]. These data together indicate that overexpressing 3R-tau isoforms results in DNA damages in adult hippocampus.

Supplement of antioxidants attenuates 3R-tau-induced DNA damages and oxidative stress

To explore whether antioxidants could improve 3R-tau-induced oxidative stress and DNA damages, we first treated primary hippocampal neurons (14 div) with VitC or VitE, the most powerful natural antioxidants [47], and measured DSB by γH2A.X staining after 24 h. VitC at 500 μM and VitE at 50 μM remarkably decreased the number of γH2A.X-positive neurons (Fig. 5a–d) and the protein level of 53BP1 (Fig. 5e, f).

Fig.5

Antioxidants attenuates 3R-tau-induced DSBs and oxidative stress. a–d) Hippocampal neurons (7 div) were infected with AAV-eGFP-3R-tau, and 7 days later the neurons were treated with VitC at 125 μM (L), 250 μM (M), and 500 μM (H), or VitE at 12.5 μM (L), 25 μM (M), and 50 μM (H) for 24 h. High concentrations of VitC or VitE decreased nuclear staining of γH2A.x (red) (One-way ANOVA, F_6,133 = 18.23, p < 0.01; Bonferroni post hoc tests, **p < 0.01 versus 3R; at least 20 cells were analyzed for each group). e, f) The mice were stereotaxically infected with the mixture of three AAV-eGFP-3R-tau isoforms on CA3 for 2 months, and VitC (100 mg/kg) or VitE (10 mg/kg) was supplemented through intraperitoneal for 14 days. Both VitC and VitE attenuated 3R-tau-induced increase of 53BP1 in hippocampal CA3 (One-way ANOVA, F_2,13 = 4.21, p < 0.05; Bonferroni post hoc tests, *p < 0.05 versus 3R; at least 4 mice for each group). g–i) Both VitC and VitE attenuated 3R-tau-induced oxidative stress evidenced by decreased MDA (g) (p = 0.0348, 3R versus Vec; p = 0.0482, 3R+VitC versus 3R; p = 0.0286, 3R+VitE versus 3R; t test; n = 6–8), increased SOD (h) (p = 0.0176, 3R versus Vec; p = 0.0476, 3R+VitC versus 3R; p = 0.0776, 3R+VitE versus 3R; t test; n = 6–8) and GSH-px (i) (p = 0.0024, 3R versus Vec; p = 0.035, 3R+VitC versus 3R; p = 0.0406, 3R+VitE versus 3R; t test; n = 6–8). Data were expressed as mean±SEM. *p < 0.05 versus 3R, **p < 0.01 versus 3R.

By intraperitoneal injection of VitC (100 mg/kg/d), VitE (10 mg/kg/d), or the vehicle into the 3R-tau mice for 14 days, we also found that VitC or VitE supplement decreased MDA level with increases of SOD and GSH-Px activity in the hippocampus of 3R-tau mice (Fig 5g, h). By using SoNar sensors that allow specific monitoring of dynamic changes of NADH level in mitochondria [48], we observed that co-expression of 3R-tau isoforms significantly decreased NADH level with an increased ratio of NAD+/NADH (Supplementary Figure 5a, b), while VitC and VitE supplement also restored NAD+/NADH ratio in HEK293T cell (Supplementary Figure 5c, d). These in vitro and in vivo data indicate that both water-soluble VitC and lipid-soluble VitE can efficiently attenuate 3R-tau-induced oxidative stress and DNA/mitochondria damages.

Supplement of antioxidants improves 3R-tau-induced memory deficit and synaptic dysfunction

Finally, we studied whether supplement of VitC or VitE could improve the 3R-tau-induced synaptic and cognitive impairments. After intraperitoneal injecting VitC, VitE, or the vehicle into 3R-tau mice for 14 days, we found that both VitC and VitE could improve spatial learning and memory deficits, though the effect of VitE was much weaker than that of VitC (Fig. 6a–h). Both VitC and VitE restored the levels of synaptic-related proteins GluA2 and GluN1 in 3R-tau mice (Fig. 6i, j). High dosage of VitC and VitE restored spine densities in cultured hippocampal neurons (Fig. 6k, l). Furthermore, VitC also inhibited activation of GFAP+-astrocytes and Iba1+-microglia cells in 3R-tau mice (Supplementary Figure 5a–d), while VitE only attenuated microglia activation without changing astrocyte number (Supplementary Figure 6a–d). Importantly, both VitC and VitE prevented neuron loss in hippocampal CA3 subset (Supplementary Figure 6e, f). These data together suggest both VitC and VitE can attenuate 3R-tau-induced toxicities, and VitC seems more effective than VitE.

Fig.6

Antioxidants rescues 3R-tau-induced cognitive deficits and synaptic deficits. The mice were stereotaxically infected with the mixture of three AAV-eGFP-3R-tau isoforms (3R) or the empty vector (Vec) on CA3 for 2 months, and VitC (100 mg/kg) or VitE (10 mg/kg) was supplemented through intraperitoneal for 2 weeks. a–d) VitC supplement restored spatial learning (a) and memory (c, d) deficits on 3R-tau mice with no effect on Vec mice (Two-way ANOVA column factor, F_3,210 = 9.88, p < 0.05; Bonferroni post hoc tests, p = 0.015, 3R+VitC versus 3R, latency; p = 0.04, 3R+VitC versus 3R, number of target platform crossing; p = 0.045, 3R+VitC versus 3R, number of effective area crossing; n = 10 mice for each group). e–h) VitE completely recovered spatial learning (e) with partially improved spatial memory ability (f–h) on 3R-tau mice, and with no positive effect on Vec mice (Two-way ANOVA column factor, F_3,220 = 9.2, p < 0.05; Bonferroni post hoc tests, p = 0.031, 3R+VitE versus 3R in latency; p = 0.11, 3R+VitE versus 3R in number of target platform crossing; p = 0.076, 3R+VitE versus 3R in number of effective area crossing; n = 10–11 mice for each group). i, j) VitC and VitE differentially promoted expression of synaptic proteins in both 3R-tau and Vec group (*p < 0.05, **p < 0.01 versus Vec, ^#p < 0.05 versus 3R; two-tail t test; n = 4 mice for each group). k, l) High dose of VitC and VitE rescued 3R-tau increased dendritic spine densities in primary hippocampal neurons cultured for 14 div (One-way ANOVA, F_5,174 = 4.724, p < 0.01; Bonferroni post hoc tests, p = 0.0011, Vec+VitC versus Vec; p = 0.0027, Vec+VitE versus Vec; p < 0.0001, 3R+VitC versus 3R; p = 0.005, 3R+VitE versus 3R; n = 30–35 for each group). Data were expressed as mean±SEM. *p < 0.05, **p < 0.01 versus Vec, ^#p < 0.05 versus 3R, ^# #p < 0.01 versus 3R.

DISCUSSION

Tau is well established as a microtubule-associated protein in neurons [1, 2]. However, under pathological conditions, aberrant assembly of tau into insoluble aggregates is accompanied by synaptic dysfunction and neural cell death in a range of neurodegenerative disorders, collectively referred to as tauopathies [49]. The formation of neurofibrillary tangles and other tau inclusion is correlated with the severity of neurodegeneration and cognitive impairment in tauopathies [50 –52], and this association is considered to continue throughout the disease course [53]. Therefore, elucidating the toxic effects and the underlying mechanisms of tau accumulation is important for understanding the pathogenesis and developing interventions for neurodegeneration.

Generated by alternative splicing of a single mRNA transcript, at least six protein isoforms of tau have been identified in the adult human brain. Tau isoforms with either three or four 32 amino acid repeats in the microtubule binding domain, denominated as 3R and 4 R Tau, respectively [8]. Tau expression is developmentally regulated, such that both 3R-tau and 4R-tau are equally expressed in normal human adult brain, while in fetal brains only 0N3R tau is expressed [6 , 54]. A number of studies show significantly increased 3R-tau/4R-tau ratio and the pathological role of 3R-tau in AD [12 , 55–57]. For instance, almost all Down syndrome patients develop typical AD pathologies by the age of 40 [58], and Down syndrome mice show increased 3R-tau and decreased 4R-tau in the brain with learning and memory deficits [59]. Moreover, only 3R-tau was predominantly detected in the tangle-bearing neuron [12, 13]. Overexpressing 0N3R tau in transgenic mice caused age-dependent pathology similar to FTDP-17 [60]. A transgenic mouse model expressing mutant 3R-tau displayed a time-dependent accumulation of 3R-tau, with toxic effects of inclusion formation, behavioral deficits, and neurodegeneration as seen in PiD [61]. Moreover, while insertion of human MAPT gene in a mouse tau KO context resulted in tau hyperphosphorylation and accumulation of insoluble 3R-tau [33], these mice also develop neurodegeneration and age-dependent memory loss [62].

Most tau related neurodegenerative diseases, such as sporadic dementia, are caused by the co-accumulation of wild-type tau isoforms, rather than by mutant tau proteins [3]. Therefore, the primary goal of the current study was to explore whether co-expressing three wild-type 3R-tau isoforms induces toxic effects as seen in the human tauopathies, and if so, the molecular mechanisms and the therapeutic strategies. To this end, we simultaneously expressed three isoforms of human wild-type 3R-tau in the hippocampus of adult mice, in which no endogenous 3R-tau protein was detected [31 –33]. We found that co-expressing 3R-tau isoforms induced tau hyperphosphorylation and formed Pick bodies, as seen in patient with PiD [63, 64], and simultaneously the mice showed learning and memory deficits with synapse impairments. We also observed that co-expressing 3R-tau in mice activated both astrocytes and microglia, which were also shown in PiD [65]. Furthermore, 3R-tau induced oxidative stress, DNA damage, and cell death, while pharmacological upregulation of antioxidants attenuated oxidative damages with amelioration of DNA damages and cognitive functions. To our best knowledge, this is the first evidence revealing the pathological role and the mechanisms of the wild-type 3R-tau isoforms, and as well as the strategies for intervention.

Studies have demonstrated that AAV- and Lenti-virus are efficient carriers for protein expression in primary neurons [66] and in the brains [67 –69]. We show that exogenous transfection could induce a dose-dependent intracellular tau accumulation [70], which allows accurately identification of the pathophysiological effects of tau. We also demonstrated by using different techniques that AAV-delivered robust co-expression of tau isoforms in neurons could mimic the PiD-like intracellular accumulation of the hyperphosphorylated tau proteins. Tau is a major neuronal microtubule-associated protein predominantly located in the neuronal axons, while level of tau was detected in astrocytes and oligodendrocytes [71, 72]. In the present study, we used CMV promoter to co-express three isoforms of 3R-tau, which might explain why AAV-CMV-tau was also detected in microglia and astrocytes. In addition, we found remarkable activation of the glial cells. However, mechanisms underlying glial activation may deserve further investigation.

Cells have developed consummate system to manage oxidative stress. Among them, the antioxidant family members, GSH-Px and SOD, can protect against the detrimental effects of ROS [51], while MDA is a product of lipid peroxidation. Several studies have reported activation of antioxidant defenses in several tauopathies [73]. In particular, SOD and GSH-Px levels were increased in some progressive supranuclear palsy patients, and were positively correlated with disease progression [74, 75]. In our study, we found that co-expressing human 3R-tau isoforms inhibited SOD and GSH-Px with elevation of MDA, demonstrate that wild-type 3R-tau protein expression can induce oxidative stress in neurons. The significantly increased oxidative stress markers have been detected in the brains of AD and the related tauopathies [76, 77]. However, oxidative stress has been shown to induce tau hyperphosphorylation [78, 79] and inhibits tau degradation [80], although its mechanism remains controversial. In this context, tau hyperphosphorylation and oxidative stress appear as two elements of a crucial “vicious circle” leading to a progressive coordinated increase in both ROS and abnormal tau. That means accumulation of 3R-tau induces oxidative stress, which in turn promotes phosphorylation of tau, forming an amplification effect.

Oxidative stress can cause DSBs that is also detected in the hippocampus of AD patients [81]. Here, we found that overexpressing 3R-tau isoforms in cultured hippocampal neurons or in adult mouse brain caused significant DSBs. In physiological conditions, DNA damage will trigger timely DNA repair to maintain cell survival. Phosphorylation of histone protein H2A.X plays a key role in early stages of cellular response to DSBs, which also contributes to cell apoptosis. Different from single 4R-tau isoform overexpressing model, in which the increased 4R-tau protein renders the cells more resistant to apoptosis [81 –84]. We observed in the current study that co-expressing three 3R-tau isoforms induced caspase-3 activation with remarkable neuron loss. It seems that the functions of tau proteins in physiology and pathology are perplexing, and different tau isoforms may have distinct effects on the particular functions of the target neurons.

We also detected synaptic and spine deficits by co-expressing 3R-tau isoforms. The synaptic damages are the structural bases for the impaired cognitive functions by 3R-tau. Although the precise mechanisms underlying tau-induced oxidative stress remain to be elucidated, we demonstrate that supplement of antioxidants, including water-soluble VitC and lipid-soluble VitE, significantly suppressed 3R-tau-induced neuropathology with cognitive improvement. The similar beneficial results were also reported by using transgenic mouse models of human tauopathies [85].

Our current data further underscore the therapeutic potential of the antioxidant for tauopathies. As mouse does not express endogenous 3R-tau, the current paper also provides in vitro (mouse primary neuron cultures) and in vivo (mouse brain) models for further investigating the pathophysiological role and the mechanisms of 3R-tau. Furthermore, simultaneous overexpressing 3R-tau isoforms could be a promising model for drug development of tauopathies, because no significant cell death has been detected in the currently available wild-type tau-accumulating models, especially at the early stage of the tauopathies.

Conclusion

Co-accumulation of 3R-tau isoforms induces memory deficit with the mechanisms involving oxidation-medicated DNA/mitochondrial damages and neuron death, and intraperitoneal administration of antioxidants VitC and VitE efficiently attenuates 3R-tau-induced pathologies and behavioral abnormalities (Supplementary Figure 7).

Footnotes

ACKNOWLEDGMENTS

We thank Dr. Fei Liu (Jiangsu Key Laboratory of Neuroregeneration) for the pEGFP-tau-0N3R, pEGFP-tau-1N3R, pEGFP-tau-2N3R, pEGFP-tau-0N4 R, pEGFP-tau-1N4 R, and pEGFP-tau-2N4 R plasmids, and Dr. Yuzheng Zhao (State Key Laboratory of Bioreactor Engineering) for SoNar sensor. This work was supported in parts by grants from Natural Science Foundation of China (31730035, 91632305, 81870846 and 81721005), from the Ministry of Science and Technology of China (2016YFC13058001), and from Sanming Project of Medicine in Shenzhen (SZSM201611090).

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Caceres

, Kosik

(1990) Inhibition of neurite polarity by tau antisense oligonucleotides in primary cerebellar neurons. Nature 343, 461–463.

Drubin

, Feinstein

, Shooter

, Kirschner

(1985) Nerve growth factor-induced neurite outgrowth in PC12 cells involves the coordinate induction of microtubule assembly and assembly-promoting factors. J Cell Biol 101, 1799–1807.

Grundke-Iqbal

, Iqbal

, Tung

, Quinlan

, Wisniewski

, Binder

(1986) Abnormal phosphorylation of the microtubule-associated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci U S A 83, 4913–4917.

Lee

, Goedert

, Trojanowski

(2001) Neurodegenerative tauopathies. Annu Rev Neurosci 24, 1121–1159.

Mandelkow

, Stamer

, Vogel

, Thies

, Mandelkow

(2003) Clogging of axons by tau, inhibition of al traffic and starvation of synapses. Neurobiol Aging 24, 1079–1085.

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.

Goedert

, Spillantini

, Potier

, Ulrich

, Crowther

(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, 393–399.

Andreadis

, Brown

, Kosik

(1992) Structure and novel exons of the human tau gene. Biochemistry 31, 10626–10633.

Lee

, Trojanowski

(1999) Neurodegenerative tauopathies: human disease and transgenic mouse models. Neuron 24, 507–510.

10.

Gotz

, Tolnay

, Barmettler

, Chen

, Probst

, Nitsch

(2001) Oligodendroglial tau filament formation in transgenic mice expressing G272V tau. Eur J Neurosci 13, 2131–2140.

11.

Gotz

, Deters

, Doldissen

, Bokhari

, Ke

, Wiesner

, Schonrock

, Ittner

(2007) A decade of tau transgenic animal models and beyond. Brain Pathol 17, 91–103.

12.

Liu

, Le

, Adamson

, Baker

, Cookson

, Hardy

, Hutton

, Yen

, Dickson

(2001) Relationship of the extended tau haplotype to tau biochemistry and neuropathology in progressive supranuclear palsy. Ann Neurol 50, 494–502.

13.

de Silva

, Lashley

, Gibb

, Hanger

, Hope

, Reid

, Bandopadhyay

, Utton

, Strand

, Jowett

, Khan

, Anderton

, Wood

, Holton

, Revesz

, Lees

(2003) Pathological inclusion bodies in tauopathies contain distinct complements of tau with three or four microtubule-binding repeat domains as demonstrated by new specific monoclonal antibodies. Neuropathol Appl Neurobiol 29, 288–302.

14.

Barker

, Luis

, Kashuba

, Luis

, Harwood

, Loewenstein

, Waters

, Jimison

, Shepherd

, Sevush

, Graff-Radford

, Newland

, Todd

, Miller

, Gold

, Heilman

, Doty

, Goodman

, Robinson

, Pearl

, Dickson

, Duara

(2002) Relative frequencies of Alzheimer disease, Lewy body, vascular and frontotemporal dementia, and hippocampal sclerosis in the State of Florida Brain Bank. Alzheimer Dis Assoc Disord 16, 203–212.

15.

Dickson

, Kouri

, Murray

, Josephs

(2011) Neuropathology of frontotemporal lobar degeneration-tau (FTLD-tau). J Mol Neurosci 45, 384–389.

16.

Cairns

, Bigio

, Mackenzie

IRA

, Neumann

, Lee

VMY

, Hatanpaa

, White

, Schneider

, Grinberg

, Halliday

, Duyckaerts

, Lowe

, Holm

, Tolnay

, Okamoto

, Yokoo

, Murayama

, Woulfe

, Munoz

, Dickson

, Ince

, Trojanowski

, Mann

DMA

(2007) Neuropathologic diagnostic and nosologic criteria for frontotemporal lobar degeneration: consensus of the Consortium for Frontotemporal Lobar Degeneration. Acta Neuropathol 114, 5–22.

17.

Bronner

, ter Meulen

, Azmani

, Severijnen

, Willemsen

, Kamphorst

, Ravid

, Heutink

, van Swieten

(2005) Hereditary Pick’s disease with the G272V tau mutation shows predominant three-repeat tau pathology. Brain 128, 2645–2653.

18.

Castellani

, Smith

, Richey

, Kalaria

, Gambetti

, Perry

(1995) Evidence for oxidative stress in Pick disease and corticobasal degeneration. Brain Res 696, 268–271.

19.

Melov

, Adlard

, Morten

, Johnson

, Golden

, Hinerfeld

, Schilling

, Mavros

, Masters

, Volitakis

, Li

, Laughton

, Hubbard

, Cherny

, Gibson

, Bush

(2007) Mitochondrial oxidative stress causes hyperphosphorylation of tau. PLoS One 2, e536.

20.

Pratico

, Delanty

(2000) Oxidative injury in diseases of the central nervous system: focus on Alzheimer’s disease. Am J Med 109, 577–585.

21.

Frost

, Hemberg

, Lewis

, Feany

(2014) Tau promotes neurodegeneration through global chromatin relaxation. Nat Neurosci 17, 357–366.

22.

Suberbielle

, Sanchez

, Kravitz

, Wang

, Ho

, Eilertson

, Devidze

, Kreitzer

, Mucke

(2013) Physiologic brain activity causes DNA double-strand breaks in neurons, with exacerbation by amyloid-beta. Nat Neurosci 16, 613–621.

23.

Tiscornia

, Singer

, Verma

(2006) Production and purification of lentiviral vectors. Nat Protoc 1, 241–245.

24.

Deacon

(2006) Assessing nest building in mice. Nat Protoc 1, 1117–1119.

25.

Denenberg

(1969) Open-field bheavior in the rat: what does it mean? Ann N Y Acad Sci 159, 852–859.

26.

Frye

, Koonce

, Edinger

, Osborne

, Walf

(2008) Androgens with activity at estrogen receptor beta have anxiolytic and cognitive-enhancing effects in male rats and mice. Horm Behav 54, 726–734.

27.

Janc

, Muller

(2014) The free radical scavenger Trolox dampens neuronal hyperexcitability, reinstates synaptic plasticity, and improves hypoxia tolerance in a mouse model of Rett syndrome. Front Cell Neurosci 8, 56.

28.

Jiang

, Chai

, Wang

, Hu

, Li

, Ma

, Wang

, Liu

(2015) CaMKII-dependent dendrite ramification and spine generation promote spatial training-induced memory improvement in a rat model of sporadic Alzheimer’s disease. Neurobiol Aging 36, 867–876.

29.

, Jiang

, Liu

, Tian

, Chen

, Yin

, Liu

, Wang

, Zhu

(2013) Reduction and the intracellular translocation of EphB2 in Tg2576 mice and the effects of beta-amyloid. Neuropathol Appl Neurobiol 39, 612–622.

30.

Zhao

, Hu

, Cheng

, Su

, Wang

, Zou

, Hu

, Chen

, Zhou

, Huang

, Yang

, Zhu

, Wang

, Yi

, Zhu

, Qian

, Chen

, Tang

, Loscalzo

, Yang

(2015) SoNar, a highly responsive NAD+/NADH sensor, allows high-throughput metabolic screening of anti-tumor agents. Cell Metab 21, 777–789.

31.

Takuma

, Arawaka

, Mori

(2003) Isoforms changes of tau protein during development in various species. Brain Res Dev Brain Res 142, 121–127.

32.

McMillan

, Korvatska

, Poorkaj

, Evstafjeva

, Robinson

, Greenup

, Leverenz

, Schellenberg

, D’Souza

(2008) Tau isoform regulation is region- and cell-specific in mouse brain. J Comp Neurol 511, 788–803.

33.

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.

34.

DeVos

, Miller

, Schoch

, Holmes

, Kebodeaux

, Wegener

, Chen

, Shen

, Tran

, Nichols

, Zanardi

, Kordasiewicz

, Swayze

, Bennett

, Diamond

, Miller

(2017) Tau reduction prevents neuronal loss and reverses pathological tau deposition and seeding in mice with tauopathy. Sci Transl Med 9, eaag0481.

35.

Benito

, Barco

(2010) CREB’s control of intrinsic and synaptic plasticity: Implications for CREB-dependent memory models. Trends Neurosci 33, 230–240.

36.

Schofield

, Kersaitis

, Shepherd

, Kril

, Halliday

(2003) Severity of gliosis in Pick’s disease and frontotemporal lobar degeneration: tau-positive glia differentiate these disorders. Brain 126, 827–840.

37.

Lopez

, Klein

, Dimbil

, Scott

(2011) Anatomically defined neuron-based rescue of neurodegenerative Niemann-Pick type C disorder. J Neurosci 31, 4367–4378.

38.

Baudry

, Yao

, Simmons

, Liu

, Bi

(2003) Postnatal development of inflammation in a murine model of Niemann-Pick type C disease: immunohistochemical observations of microglia and astroglia. Exp Neurol 184, 887–903.

39.

Sheffield

, Marquis

, Berman

(2000) Regional distribution of cortical microglia parallels that of neurofibrillary tangles in Alzheimer’s disease. Neurosci Lett 285, 165–168.

40.

Grundke-Iqbal

, Fleming

, Tung

, Lassmann

, Iqbal

, Joshi

(1990) Ferritin is a component of the neuritic (senile) plaque in Alzheimer dementia. Acta Neuropathol 81, 105–110.

41.

Markesbery

(1997) Oxidative stress hypothesis in Alzheimer’s disease. Free Radic Biol Med 23, 134–147.

42.

Nunomura

, Perry

, Aliev

, Hirai

, Takeda

, Balraj

, Jones

, Ghanbari

, Wataya

, Shimohama

, Chiba

, Atwood

, Petersen

, Smith

(2001) Oxidative damage is the earliest event in Alzheimer disease. J Neuropathol Exp Neurol 60, 759–767.

43.

Kim

, Frank

, Dobbin

, Tsunemoto

, Tu

, Peng

, Guan

, Lee

, Moy

, Giusti

, Broodie

, Mazitschek

, Delalle

, Haggarty

, Neve

, Lu

, Tsai

(2008) Deregulation of HDAC1 by p25/Cdk5 in neurotoxicity. Neuron 60, 803–817.

44.

Bennett

, Lewis

, Baldwin

, Resnick

(1993) Lethality induced by a single site-specific double-strand break in a dispensable yeast plasmid. Proc Natl Acad Sci U S A 90, 5613–5617.

45.

Sandell

, Zakian

(1993) Loss of a yeast telomere: arrest, recovery, and chromosome loss. Cell 75, 729–739.

46.

Panier

, Boulton

(2014) Double-strand break repair: 53BP1 comes into focus. Nat Rev Mol Cell Biol 15, 7–18.

47.

Tucker

, Townsend

(2005) Alpha-tocopherol: roles in prevention and therapy of human disease. Biomed Pharmacother 59, 380–387.

48.

Zhao

, Jin

, Hu

, Zhou

, Yi

, Yu

, Xu

, Wang

, Yang

, Loscalzo

(2011) Genetically encoded fluorescent sensors for intracellular NADH detection. Cell Metab 14, 555–566.

49.

Guo

, Noble

, Hanger

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

50.

Arriagada

, Growdon

, Hedley-Whyte

, Hyman

(1992) Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology 42, 631–639.

51.

Alavi Naini

, Soussi-Yanicostas

(2015) Tau hyperphosphorylation and oxidative stress, a critical vicious circle in neurodegenerative tauopathies? Oxid Med Cell Longev 2015, 151979.

52.

Bierer

, Hof

, Purohit

, Carlin

, Schmeidler

, Davis

, Perl

(1995) Neocortical neurofibrillary tangles correlate with dementia severity in Alzheimer’s disease. Arch Neurol 52, 81–88.

53.

Abner

, Kryscio

, Schmitt

, Santacruz

, Jicha

, Lin

, Neltner

, Smith

, Van Eldik

, Nelson

(2011) “End-stage” neurofibrillary tangle pathology in preclinical Alzheimer’s disease: fact or fiction? J Alzheimers Dis 25, 445–453.

54.

Kosik

, Orecchio

, Bakalis

, Neve

(1989) Developmentally regulated expression of specific tau sequences. Neuron 2, 1389–1397.

55.

Jin

, Yin

, Gu

, Zhang

, Shi

, Qian

, Ji

, Cao

, Gu

, Ding

, Iqbal

, Gong

, Liu

(2015) Truncation and activation of dual specificity tyrosine phosphorylation-regulated kinase 1A by Calpain I: a molecular mechanism linked to tau pathology in Alzheimer disease. J Biol Chem 290, 15219–15237.

56.

Espinoza

, de Silva

, Dickson

, Davies

(2008) Differential incorporation of tau isoforms in Alzheimer’s disease. J Alzheimers Dis 14, 1–16.

57.

Connell

, Rodriguez-Martin

, Gibb

, Kahn

, Grierson

, Hanger

, Revesz

, Lantos

, Anderton

, Gallo

(2005) Quantitative analysis of tau isoform transcripts in sporadic tauopathies. Brain Res Mol Brain Res 137, 104–109.

58.

Wisniewski

, Wisniewski

, Wen

(1985) Occurrence of neuropathological changes and dementia of Alzheimer’s disease in Down’s syndrome. Ann Neurol 17, 278–282.

59.

Yin

, Jin

, Shi

, Zhang

, Wu

, Gong

, Iqbal

, Liu

(2017) Dyrk1A overexpression leads to increase of 3R-tau expression and cognitive deficits in Ts65Dn Down syndrome mice. Sci Rep 7, 619.

60.

Ishihara

, Hong

, Zhang

, Nakagawa

, Lee

, Trojanowski

, Lee

(1999) Age-dependent emergence and progression of a tauopathy in transgenic mice overexpressing the shortest human tau isoform. Neuron 24, 751–762.

61.

Rockenstein

, Overk

, Ubhi

, Mante

, Patrick

, Adame

, Bisquert

, Trejo-Morales

, Spencer

, Masliah

(2015) A novel triple repeat mutant tau transgenic model that mimics aspects of Pick’s disease and fronto-temporal tauopathies. Plos One 10, e0121570.

62.

Polydoro

, Acker

, Duff

, Castillo

, Davies

(2009) Age-dependent impairment of cognitive and synaptic function in the htau mouse model of tau pathology. J Neurosci 29, 10741–10749.

63.

Ferrer

, Barrachina

, Puig

(2002) Glycogen synthase kinase-3 is associated with neuronal and glial hyperphosphorylated tau deposits in Alzheimer’s disease, Pick’s disease, progressive supranuclear palsy and corticobasal degeneration. Acta Neuropathol 104, 583–591.

64.

Ransohoff

(2016) How neuroinflammation contributes to neurodegeneration. Science 353, 777–783.

65.

Dickson

(2001) Neuropathology of Pick’s disease. Neurology 56, S16–S20.

66.

Hughes

, Moussavi-Harami

, Sauter

, Davidson

(2002) Viral-mediated gene transfer to mouse primary neural progenitor cells. Mol Ther 5, 16–24.

67.

Burger

, Nash

, Mandel

(2005) Recombinant adeno-associated viral vectors in the nervous system. Hum Gene Ther 16, 781–791.

68.

Kaplitt

, Leone

, Samulski

, Xiao

, Pfaff

, O’Malley

, During

(1994) Long-term gene expression and phenotypic correction using adeno-associated virus vectors in the mammalian brain. Nat Genet 8, 148–154.

69.

Kay

, Manno

, Ragni

, Larson

, Couto

, McClelland

, Glader

, Chew

, Tai

, Herzog

, Arruda

, Johnson

, Scallan

, Skarsgard

, Flake

, High

(2000) Evidence for gene transfer and expression of factor IX in haemophilia B patients treated with an AAV vector. Nat Genet 24, 257–261.

70.

, Li

, Wang

, Luo

, Zhang

, Liu

, Feng

, Wang

, Yue

, Chen

, Ye

, Wang

, Liu

(2016) Tau accumulation impairs mitophagy via increasing mitochondrial membrane potential and reducing mitochondrial Parkin. Oncotarget 7, 17356–17368.

71.

Kempf

, Clement

, Faissner

, Lee

, Brandt

(1996) Tau binds to the distal axon early in development of polarity in a microtubule- and microfilament-dependent manner. J Neurosci 16, 5583–5592.

72.

Klein

, Kramer

, Cardine

, Schraven

, Brandt

, Trotter

(2002) Process outgrowth of oligodendrocytes is promoted by interaction of fyn kinase with the cytoskeletal protein tau. J Neurosci 22, 698–707.

73.

Markesbery

, Lovell

(1998) Four-hydroxynonenal, a product of lipid peroxidation, is increased in the brain in Alzheimer’s disease. Neurobiol Aging 19, 33–36.

74.

Aoyama

, Matsubara

, Kobayashi

(2006) Aging and oxidative stress in progressive supranuclear palsy. Eur J Neurol 13, 89–92.

75.

Cantuti-Castelvetri

, Keller-McGandy

, Albers

, Beal

, Vonsattel

, Standaert

, Augood

(2002) Expression and activity of antioxidants in the brain in progressive supranuclear palsy. Brain Res 930, 170–181.

76.

Smith

, Richey Harris

, Sayre

, Beckman

, Perry

(1997) Widespread peroxynitrite-mediated damage in Alzheimer’s disease. J Neurosci 17, 2653–2657.

77.

Bruce-Keller

, Li

, Lovell

, Kraemer

, Gary

, Brown

, Markesbery

, Mattson

(1998) 4-Hydroxynonenal, a product of lipid peroxidation, damages cholinergic neurons and impairs visuospatial memory in rats. J Neuropathol Exp Neurol 57, 257–267.

78.

Davis

, Anderton

, Brion

, Reynolds

, Hanger

(1997) Oxidative stress induces dephosphorylation of tau in rat brain primary neuronal cultures. J Neurochem 68, 1590–1597.

79.

, Odawara

, Yen

(1997) Menadione-induced tau dephosphorylation in cultured human neuroblastoma cells. Brain Res 760, 118–128.

80.

Guttmann

, Johnson

(1998) Oxidative stress inhibits calpain activity in situ. J Biol Chem 273, 13331–13338.

81.

Myung

, Zhu

, Kruman

, Castellani

, Petersen

, Siedlak

, Perry

, Smith

, Lee

(2008) Evidence of DNA damage in Alzheimer disease: phosphorylation of histone H2AX in astrocytes. Age (Dordr) 30, 209–215.

82.

Wang

, Li

, Liu

, Zhang

, Liao

, Wang

, Liu

(2010) Tau overexpression inhibits cell apoptosis with the mechanisms involving multiple viability-related factors. J Alzheimers Dis 21, 167–179.

83.

Wang

, Yin

, Zhang

, Zhu

, Tian

, Wang

, Li

, Wang

(2010) Overexpression of tau proteins antagonizes amyloid-beta-potentiated apoptosis through mitochondria-caspase-3 pathway in N2a cells. J Alzheimers Dis 20, 145–157.

84.

Utton

, Gibb

, Burdett

, Anderton

, Vandecandelaere

(2001) Functional differences of tau isoforms containing 3 or 4 C-terminal repeat regions and the influence of oxidative stress. J Biol Chem 276, 34288–34297.

85.

Nakashima

, Ishihara

, Yokota

, Terada

, Trojanowski

, Lee

, Kuroda

(2004) Effects of alpha-tocopherol on an animal model of tauopathies. Free Radic Biol Med 37, 176–186.