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

Learn More →

Equilibrative nucleoside transporter 1 inhibition rescues energy dysfunction and pathology in a model of tauopathy

Equilibrative nucleoside transporter 1 inhibition rescues energy dysfunction and pathology in a... Tau pathology is instrumental in the gradual loss of neuronal functions and cognitive decline in tauopathies, including Alzheimer’s disease (AD). Earlier reports showed that adenosine metabolism is abnormal in the brain of AD patients while consequences remained ill‑ defined. Herein, we aimed at investigating whether manipulation of adenosine tone would impact Tau pathology, associated molecular alterations and subsequent neurodegeneration. We demonstrated that treatment with an inhibitor (J4) of equilibrative nucleoside transporter 1 (ENT1) exerted beneficial effects in a mouse model of Tauopathy. Treatment with J4 not only reduced Tau hyperphosphorylation but also rescued memory deficits, mitochondrial dysfunction, synaptic loss, and abnormal expression of immune ‑related gene signatures. These beneficial effects were particularly ascribed to the ability of J4 to suppress the overactivation of AMPK (an energy reduction sensor), suggesting that normalization of energy dysfunction mitigates neuronal dysfunctions in Tauopathy. Collectively, these data highlight that targeting adenosine metabolism is a novel strategy for tauopathies. Keywords: Alzheimer’s disease, Tauopathy, Adenosine, AMPK, ENT1 Background 30 kinases and protein phosphatases [6, 7], thus making Alzheimer’s disease (AD) is the most prominent neuro- it very sensitive to the environment including metabolic degenerative disease in aging societies, but there is no changes [8, 9]. effective treatment [1]. The major pathogenic hallmarks Adenosine is an important homeostatic building block of Alzheimer’s disease include extracellular amyloid of many important metabolic pathways. It also serves as a plaques (amyloid-beta, Aβ) and intracellular accumula- neuromodulator that controls multiple functions (includ- tion of neurofibrillary tangles made of hyperphosphoryl - ing neuroinflammation, blood–brain barrier  permeabil - ated Tau. The latter is particularly known to be associated ity, neuronal transmission, and energy balance) through with neuritic dystrophy, synapse loss, and neuroinflam - receptor-dependent and/or independent mechanisms in mation, which lead to cognitive impairments [2–5]. the central nervous system [10, 11]. The main sources of Phosphorylation of Tau can be modulated by more than adenosine include ATP catabolism and the transmethyla- tion pathway. In addition, the extracellular and intracellu- lar adenosine levels are modulated through equilibrative *Correspondence: david.blum@inserm.fr; bmychern@ibms.sinica.edu.tw nucleoside transporters (ENTs) and concentrative nucle- Institute of Biomedical Sciences, Academia Sinica, Nankang, Taipei 115, Taiwan oside transporters [12, 13]. ENTs are bidirectional Univ. Lille, Inserm, CHU Lille, U1172 ‑ LilNCog ‑ Lille Neuroscience & transporter that transports adenosine in a concentra- Cognition, 59000 Lille, France tion-dependent, Na -independent manner. These ENT Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Chang et al. acta neuropathol commun (2021) 9:112 Page 2 of 18 members contain 11 transmembrane domains, and adenosine homeostasis modulation by J4 in a Tauopa- can be found in most cell types (including neurons and thy context, using the THY-Tau22 (Tau22) model, which astrocytes) in the brain. Among the four ENTs, ENT1 progressively develops hippocampal Tau pathology and has attracted much attention in recent years because it memory deficits [33]. We demonstrated that chronic has the highest affinity for adenosine [14, 15]. Since dis- treatment with J4 mitigates the development of hip- ruption of adenosine homeostasis in the brain has been pocampal Tau pathologies (including synapse loss, mito- implicated in several neurological disorders (including chondrial dysfunction, and neuroinflammation). These sleep disorders, impaired cognition impairment and neu- beneficial effects of J4 are particularly ascribed to its abil - rodegenerative diseases [16, 17]), modulation of adeno- ity to suppress AMPK overactivation in the hippocampi sine levels by controlling the components of adenosine of Tau22 mice, suggesting that it normalizes energy dys- metabolism therefore is a potential therapeutic approach. function caused by pathogenic Tau. Impaired adenosine homeostasis has been also sug- gested in the brain of AD patients [18]. In this study, Methods Alonso-Andres et  al. had reported that the adenosine Animals and drug administration levels in the frontal cortices of AD patients were lower THY-Tau22 (Tau22) mice ((B6.Cg-Tg(y1-Th than those of age-matched controls throughout disease MAPT)22Schd) expressing mutant human 4R Tau progression. Conversely, the adenosine levels of parietal (G272V and P301S) driven by a neuron-specific pro - cortices gradually become higher than normal subjects at moter (y1.2) Th were maintained on the C57BL/6J back - the later stage of AD [18]. No data of the hippocampal ground [33]. All animal studies were conducted following adenosine level was reported in this study. These findings the protocols approved by the Institutional Animal Care overall suggest that adenosine homeostasis may be dys- and Utilization Committee (IACUC, Academia Sinica, regulated in AD in a brain area-specific manner. Accord - Taiwan). All mice were housed in ventilated cages (IVC) ingly, dysregulation of the two main adenosine receptors, with freely accessible water and chow diet (LabDiet , A and A have been reported in the brain of patients San Antonio, TX, USA), and kept under controlled light- 1 2A [19, 20] as well as in AD mouse models [21–24]. ing condition with 12:12  h light/dark cycle at the Insti- The link between adenosine homeostasis and Tau tute of Biomedical Sciences Animal Care Facility (Taipei, pathology remains ill-defined. Interestingly, previous Taiwan). To investigate the effect of J4, male Tau22 mice studies have associated AMPK activation with abnor- and their littermate controls were randomly allocated to mal Tau phosphorylation in the brains of patients and experimental groups that treated with the J4 (0.02 mg/ml mice with Alzheimer’s disease [25–28]. AMP-activated in 1% HPβCD; designated TauJ and WTJ mice) or vehi- protein kinase (AMPK) is a homeostatic energy sensor cle (1% HPβCD; designated TauC and WTC mice) con- that controls the balance between anabolic and catabolic tinuously in drinking water for 7  months from the age processes in cells [29]. In the presence of stress (e.g., of 3 to 10  months. Males were chosen for the following an elevated AMP/ATP ratio, high reactive oxygen spe- experiments. The behavioral tests, electrophysiological cies levels, or mitochondrial dysfunction), AMPK can study, and RNA-seq analysis were carried out at the age be activated by being phosphorylated at Thr on the of 10  months. The immunohistochemical staining and α subunit of AMPK [30]. Under adverse conditions, for quantitative PCR were performed at age of 11  months. instance when the extracellular adenosine concentration The phospho-proteomic analysis was assessed at age of is elevated, transport of adenosine into cells enhances 12 months. Mice had continuously received the indicated the cellular level of AMP, alters the AMP/ATP ratio, and treatment during experiments. subsequently activates AMPK [31, 32]. Together, these observations raised the possibility of a link between Human cases adenosine homeostasis, AMPK and Tau. For immunofluorescence analysis, a total of 18 post-mor - We have addressed this link using the in-house syn- tem Human posterior hippocampal specimens: six nor- thetic compound J4. J4 is an orally active, BBB-perme- mal subjects, six Alzheimer’s disease, and six FTD-Tau able inhibitor of ENT1 with a Ki value of 0.05  μM [22]. (CBD, PSP, and Pick’s), were obtained from the UC Davis Intrahippocampal acute infusion of J4 elevated the extra- Alzheimer’s Disease Center (USA). cellular adenosine level determined by microdialysis in For mRNA analysis, a total of 55 post-mortem Human THY-Tau22 and WT mice (Additional file  1: Fig. S1). Its brain samples (Brodmann area 10 prefrontal cortex or oral bioavailability is 48% in mice with brain-to-blood temporal cortex): 13 normal subjects, 19 Alzheimer’s ratio of approximately 16.4% (Additional file  1: Table S1), Disease, and 23 FTD-Tau (CBD, P301L, PSP, and Pick’s), indicating that J4 can enter the brain via oral administra- were obtained from the brain banks of Lille, Paris, and tion. In the present study, we investigated the impact of Geneva. Participants and methods have been described Chang  et al. acta neuropathol commun (2021) 9:112 Page 3 of 18 previously [34, 35]. Fresh frozen grey matter tissue (about cerebrospinal fluid (ACSF; 119  mM NaCl, 2.5  mM KCl, 100  mg) retrieved at autopsy and stored at − 80  °C was 2.5 mM CaCl , 1.3  mM M gSO , 1  mM N aH PO , 2 4 2 4 used for mRNA analysis. All the brain samples used for 26.2 mM NaHCO , and 11 mM glucose) oxygenated with RT-qPCR analyses had an RNA integrity number ≥ 5. 95% O and 5% CO . Transverse hippocampal slices of 2 2 Detailed information on the normal subjects, Alzhei- 450 μm thickness were prepared with a DSK Microslicer mer’s Disease patients, and FTD-Tau patients from which (DTK-1000, Osaka, Japan) filled with oxygenated ice- the specimens used in the present study were obtained is cold ACSF. For recovery, the slices were then incubated listed in Additional file 1: Table S2. in an interface-type holding chamber filled with oxygen - ated ACSF at RT for at least 3  h. Before recording, the Morris water maze recording electrodes were prepared with a glass micro- Spatial memory and cognitive flexibility of mice aged pipette puller (PC-10, Narishige, Tokyo, Japan), and the 10–11  months were evaluated using the Morris water slices were transferred to an immersion-type recording maze test as described with slight modifications [36]. A chamber equipped with a perfusion system (flow rate: circular swimming pool (154  cm in diameter, 51  cm in 2–3  ml/min) and temperature controller (kept at 32  °C). height) was filled with milky water (30 cm in depth, kept To record field excitatory postsynaptic potentials (fEP - at 20  °C), and divided into four quadrants (T: target; L: SPs) recording, the bipolar stainless-steel stimulating target left; R: target right; O: opposite) with distinct electrodes (Frederick Haer Company, Bowdoinham, ME; visual cues on the tank wall of each quadrant. The hid - 10 ΩM impedance) and a glass pipette filled with 3  M den platform (13  cm in diameter, 0.5  cm below the sur- NaCl were placed in the stratum radiatum of the hip- face of the milky water) was placed in the center of the pocampal CA1 region. Basal synaptic transmission at target quadrant (T). Each mouse underwent four daily the Schaffer collateral-CA1 synapses was first evaluated training trial (120 s/trial, 30 min interval), in which they by measuring input–output curves using 12 stimuli (con- were released from randomly selected nontarget quad- stant current pulses from 10 to 120 μA in increments of rants (NTs). For the spatial memory test in the acquisi- 10 μA, duration of 40  μs). To measure the paired-pulse tion-learning phase, learning trials were performed with facilitation (PPF) response, two pulses were applied in a hidden platform for five consecutive days (Day 1–Day rapid succession (interpulse intervals of 50, 100, 150, 200, 5). For the spatial reversal memory test in the reversal- 300, 400 and 500 ms). Baseline responses were recorded learning phase, learning trials were performed with a by applying single stimuli (40 μs pulse-width) at 30  s hidden platform relocated to the opposite quadrant for intervals, and 4 responses were averaged to obtain a data an additional four consecutive days (Day 9–Day 12). To point. Long-term depression (LTD) was induced using 3 evaluate reference memory, the probe trial and rever- trains of low-frequency stimulation (LFS, 1200 pulses at sal probe trial were performed on Day 8 (72  h after the 2 Hz) with a 10-min interval/train as described elsewhere acquisition-learning phase) and Day 15 (72  h after the [37]. The initial fEPSP slope was calculated by using Sig - reversal-learning phase), respectively. For the probe test, nal software (V4.08, Cambridge Electronic Design, Cam- the hidden platform was first removed. The mouse was bridge, UK). then released in the opposite quadrant (O), and their swimming path was recorded for 120  s. The swimming Phospho‑proteomic analysis path and other parameters (e.g. escape latency and swim- Hippocampal lysates (200  μg) collected from three ani- ming speed) of each mouse in different quadrants were mals from each condition were subjected to phospho- monitored and analyzed using the TrackMot video track- proteomic analysis at the Proteomics Core Facility (PCF) ing system (Diagnostic & Research Instruments Co., Ltd., (Institute of Biomedical Sciences, Academia Sinica, Tai- Taoyuan, Taiwan). Mice that exhibited nonsearching pei, Taiwan). For in-solution digestion, 8  M urea was behaviors (floating, a swimming speed below 10  cm/s added to the lysates to prepare a mixture of 10  mg pro- and circling) were excluded from the analysis. Statistical tein/ml in 6  M urea. Protein reduction was performed differences were analyzed by two-way ANOVA. by the addition of dithiothreitol (DTT, final concentra - tion 5  mM) and incubation at 56  °C for 25  min. Protein Electrophysiological study alkylation was performed by the addition of iodoaceta- Mice aged 10–11  months were used for electrophysi- mide (IAA, final concentration 15  mM) and incubation ology approaches. All electrophysiology studies were at RT for 45  min to block the reversion of sulfhydryl (– performed at the electrophysiology core facility (Neu- SH) groups to disulfide bonds. Trypsin/Lys-C Mix (Pro - roscience Program of Academia Sinica, Taipei, Tai- mega, WI, USA) was then added to protein (protein: wan). After rapid decapitation, the hippocampus was protease = 25:1 w/w) and the mixture was incubated for quickly dissected out and immersed in ice-cold artificial 4 h at 37 °C. The urea concentration was adjusted to 1 M Chang et al. acta neuropathol commun (2021) 9:112 Page 4 of 18 or less by diluting the reaction mixture with triethylam- normalization. References of the probes used in this monium bicarbonate (TEAB, 50  mM) followed by an study are given in Additional file 1: Table S4. incubation at 37  °C for 17  h. The digested samples were dried with a SpeedVac and desalted with C18 Oasis RNA sequencing (RNA‑seq) PRiME HLB cartridges (Waters, MA, USA). For iTRAQ Total RNA samples (3 μg per sample) extracted from the labeling, the digested peptides were labeled with four hippocampus with RIN values greater than 8 were sub- isobaric iTRAQ Reagents (114, 115, 116, and 117) using jected to RNA-seq analysis. The RNA library construc - iTRAQ Reagents-4plex Applications Kit (AB Sciex, tion and sequencing were carried out by Welgene Biotech MA, USA) following manufacturer’s instructions. For (Taipei, Taiwan). Briefly, the SureSelect Strand-Specific phosphopeptide enrichment, iTRAQ labeled peptides RNA Library Preparation Kit (Agilent Technology, CA, were mixed with loading buffer (80% acetonitrile, ACN, USA) was used for library construction on the Illumina 5% trifluoroacetic acid, TFA, and 1  M glycolic acid) and platform. After AMPure XP Bead-based (Beckman adjusted to pH 2. The sample solution was then mixed Coulter Genomics, MA, USA) size selection of the RNA with TiO beads (GL Sciences, Japan) and incubated with 2 library, the sequences were determined using the Illumi- vortexing at RT for 15 min. The beads were collected and na’s sequencing-by-synthesis (SBS) technology to obtain washed twice with 100 μl washing buffer (80% ACN and 150-bp paired-end reads. Sequencing data (FASTQ files) 5% TFA). The phosphopeptides were sequentially eluted were generated by Welgene’s pipeline (Base call conver- with 50 μl 0.5% N H OH, 50 μl 5% N H OH, and 50 μl 80% 4 4 sion, adaptor clipping, and sequence quality trimming) ACN with 0.1% formic acid, and dried with a SpeedVac. based on Illumina’s base-calling program bcl2fastq v2.2.0 The phosphopeptides were then subjected to LC/MS/MS (Illumina, CA, USA) and Trimmomatic v0.36 [38]. The and analyzed by Proteome Discoverer ver.2.2 (Thermo RNA-seq reads were then aligned to the mouse reference Fisher Scientific, Waltham, MA, USA). genome (mm10) from the Ensembl database (Ensembl release 93) using HISAT2 [39]. The sample-to-sample distances were visualized by principal components anal- RNA extraction, cDNA synthesis, and quantitative PCR ysis (PCA) (Additional file  1: Fig. S2). Expression levels For mouse brain tissue, RNA isolation and complemen- (fragments per kilobase per million, FPKM) were ana- tary DNA (cDNA) synthesis were performed according lyzed and estimated using cuffdiff (cufflinks v2.2.1) [40] to the manufacturer’s protocols. In brief, mouse hip- and Welgene in-house programs. To identify the differ - pocampal tissues were homogenized in GENEzol rea- entially expressed (DE) genes in different groups, cutoff gent (GZX100, Geneaid Biotech Ltd., New Taipei City, criteria (absolute lo g fold change ≥ 0.32, p < 0.05) were Taiwan) with sterilized tissue grinders (FocusBio, Tai- used. A volcano plot and heatmap of the DE genes were wan), and then standard procedures for RNA preparation drawn by using Instant Clue [41] and Morpheus software and cDNA synthesis were performed as described previ- (https:// softw are. broad insti tute. org/ morph eus/), respec- ously [22]. Quantitative PCR (qPCR) assays were carried tively. The Gene Ontology (GO) and the Kyoto Encyclo - out using the LightCycler 480 System (Roche Life Sci- pedia of Genes and Genomes (KEGG) pathways of DE ence, Indiana, USA) and analyzed by the comparative CT genes were analyzed by using the Database for Annota- (ΔΔCt) method with GAPDH as a reference gene. The tion, Visualization and Integrated Discovery (DAVID 6.7) sequences of the PCR primers are shown in Additional [42, 43]. The Ingenuity Pathway Analysis (IPA) software file 1: Table S3. (Qiagen, CA, USA) was then used for the identification For Human brain tissue, total mRNA was extracted and of canonical pathway(s) related to the DE genes. Z-score purified using the RNeasyLipid Tissue Mini Kit (Qiagen). was used to predict activation (Z-score ≥ 2.0) or inhibi- One microgram of total mRNA was reverse-transcribed tion (Z-score ≤ − 2.0) of the indicated pathway. using the HighCapacity cDNA reverse transcription kit (Applied Biosystems). Quantitative real-time polymer- Immunohistochemical staining ase chain reaction (qPCR) analysis was performed on Coronal brain sections  (20  μm) from the desired mice ™ ™ an Applied Biosystems StepOnePlus Real-Time PCR were prepared as previously described [22]. For immu- Systems using TaqMan Gene Expression Master Mix nofluorescence (IF) staining, brain slices were washed (Applied Biosystems ). The thermal cycler conditions with 0.1  M PBS buffer, permeabilized with 0.2% Triton were as follows: 95 °C for 10 min, then 40 cycles at 95 °C X-100 solution (in 0.1  M PBS buffer), and blocked with for 15  s and 60  °C for 1  min. Amplifications were car - 3% normal goat serum (NGS), 3% normal donkey serum ried out in duplicate and the relative expression of target (NDS), or 3% bovine serum albumin (BSA) in 0.1 M PBS genes was determined by the ΔΔCt method with β-Actin buffer for 2 h at RT. The brain sections were then washed (ACTB) was used as a reference housekeeping gene for Chang  et al. acta neuropathol commun (2021) 9:112 Page 5 of 18 with 0.1  M PBS buffer twice and incubated with the deaminase (ADA), CD39, CD73) involved in adeno- indicated primary antibodies (listed in Additional file  1: sine metabolism were found elevated in the hippocam- Table S5) in primary antibody solution (1% NGS or BSA, pus of Tau22 mice, supporting a change in adenosine 0.2% Triton X-100, and 0.1% sodium azide in 0.1 M PBS) homeostasis. A trend of increase in the transcript level for 48  h at 4  °C. After extensive washes, the brain sec- of adenosine kinase (ADK) was also found, but did not tions were incubated with the corresponding secondary reach statistical significance. Importantly, J4 treatment antibody (1:500) for 2  h at RT, and then the nuclei were normalized all these changes (Table  1), suggesting that stained with Hoechst 33258 (1:5000) for 10  min at RT. J4 was able to reinstate proper adenosine homeostasis Free-floating brain sections were mounted on the silane- in the hippocampus of Tau22 mice. coated slides (Muto Pure Chemicals Co., Tokyo, Japan) with mounting media (Vector Laboratories, CA, USA) and stored at 4 °C before imaging. An LSM 780 confocal Chronic J4 treatment prevents impairment of spatial microscope (Carl Zeiss, Germany) was used to capture learning and memory of Tau22 mice images. The images were analyzed with MetaMorph soft - Under the tested condition, the spatial memory and ware (Universal Imaging, PA, USA). cognitive flexibility of mice at the age of 10  months Formalin-fixed, paraffin-embedded human brain slices were evaluated using the Morris water maze (MWM) were deparaffinized and rehydrated. To expose the anti - task. J4 significantly prevented spatial learning defi - genic sites, the brain slices were then immersed in 1X cits in vehicle-treated Tau22 mice (TauC) during the citrate buffer (C9999, Sigma-Aldrich, St. Louis, MO, acquisition-learning phase (Day 1–Day 5) (Fig.  1a) USA) for 20  min at 97.5  °C and cooled to RT. The brain without affecting wild-type mice. Moreover, vehicle-, slices were subjected to immunofluorescence staining J4-treated WT, and J4-treated Tau22, but not vehicle- as described above. After secondary antibody (1:500) treated Tau22 mice showed a preference for searching incubation, the brain sections were treated with 0.1% the hidden platform in the target quadrant (T) over the (w/v) Sudan Black B (199664, Sigma) in 70% ethanol for nontarget quadrants (NTs) in the probe test (Fig.  1b), 15 min at RT to block autofluorescence signal. The brain indicating that J4 improved the impaired memory of sections were then stained with Hoechst 33258 (1:5000) Tau22 mice. for 10  min at RT. After extensive washes with PBS, the To assess cognitive flexibility, we also examined the brain sections were mounted with the mounting media spatial reversal learning of WT and Tau22 mice treated (Vector Laboratories, CA, USA) and stored at 4 °C before with or without J4 for an additional four consecutive imaging. days. In the reversal-learning phase (Day 9–Day 12), the hidden platform was relocated to the opposite quadrant and the escape latency was recorded. J4 improved the Statistics impaired spatial reversal learning of TauC mice (Fig. 1a). The experimental condition was blinded to investigators Animals were subjected to the probe test on Day 15. during behavioral and electrophysiological experiments. Except for vehicle-treated Tau22 mice, all mice spent The data are expressed as the mean ± S.E.M. All statisti- more time in the new target quadrant (T) than in the cal analyses were performed using GraphPad Prism Soft- nontarget quadrants (NTs) (Fig. 1c). No difference in the ware (La Jolla, CA, USA). Two-tailed unpaired Student’s swimming speed was observed among the groups tested t-test was used to compare the difference between the two groups. One-way or two-way ANOVA followed by Tukey’s multiple comparisons test was used for compari- son of multiple groups. Differences were considered sta - tistically significant when p < 0.05. Table 1 Enzymes involved in adenosine homeostasis Gene WTC WTJ TauC TauJ Results Chronic J4 treatment stabilizes the abnormally altered # ADA 1.06 ± 0.09 1.04 ± 0.06 3.47 ± 0.42* 1.35 ± 0.24 expressions of genes that control adenosine metabolism # ADK 1.02 ± 0.03 1.06 ± 0.02 1.17 ± 0.06 0.85 ± 0.09 WT and Tau22 male mice were treated with J4 # CD39 1.01 ± 0.03 1.23 ± 0.02* 1.45 ± 0.06* 0.94 ± 0.05 (0.02  mg/ml in 1% HPβCD) or vehicle (1% HPβCD) in # CD73 1.02 ± 0.04 1.23 ± 0.07* 1.33 ± 0.05* 1.13 ± 0.05 drinking water for 7 months from the age of 3 months. Mice were treated as indicated (control WT mice, WTC; J4-treated WT mice, The average daily intake level of J4 was 3.06 ± 0.28  mg/ WTJ; control Tau22 mice, TauC; and J4-treated Tau22 mice, TauJ; n = 6–9 in kg. The hippocampi of the treated animals were then each group) from the age of 3–11 months. The hippocampus was harvested carefully and subjected to RT-qPCR. GAPDH was used as a reference gene for collected to harvest total RNA for RT-qPCR analy- normalization. The data are expressed as the mean ± SEM. *p < 0.05 versus the sis. The levels of at least three genes (i.e., adenosine WT vehicle group; p < 0.05 versus the Tau22 vehicle group Chang et al. acta neuropathol commun (2021) 9:112 Page 6 of 18 (Fig.  1d). Together, J4 significantly improves the spatial Alzheimer’s Disease was further assessed by Western memory and spatial reversal memory of Tau22 mice. blot and immunofluorescence analysis using antibodies raised against hyperphosphorylated (pThr181, pSer199, Chronic J4 treatment normalizes the impairment pSer202/Thr205 = AT8, pSer262, pSer396, and pSer422) of hippocampal CA1 LTD in Tau22 mice and misfolded (pThr212/Ser214 = AT100; MC1) Tau. We next investigated whether J4 affects the synaptic plas - Chronic treatment with J4 significantly reduced Tau ticity in Tau22 mice. The basal transmission of the hip - phosphorylation at the sites tested except for pSer181 pocampal CA3–CA1 network was determined based (Figs.  2b, c, Additional file  1: S3A, and S3B). No effect on the input–output relationship. The vehicle-treated of J4 on the human Tau was observed, demonstrating Tau22 mice showed decreased basal synaptic transmis- that J4 did not modulate the Thy-1 promoter directly sion compared to that of the vehicle-treated WT mice. (Additional file  1: Fig. S3C–S3G). Importantly, the lev- Such abnormal synaptic transmission was alleviated by els of misfolded Tau, assessed by AT100 and MC1 [44, J4 in Tau22 mice (Fig.  1e). Presynaptic neurotransmitter 45], were also reduced by J4 (Fig.  2b, c). Collectively, J4 release in the hippocampus, as determined by the paired- reduces the levels of hyperphosphorylated and misfolded pulse facilitation (PPF) response assay, was comparable Tau in Tau22 hippocampi. between genotypes (Tau22 versus WT mice) and treat- ment groups (J4 versus vehicle) (Fig.  1f ), suggesting that Chronic J4 treatment reduces the AMPK activation presynaptic plasticity was unaffected in Tau22 mice. in the hippocampus of Tau22 mice Previous studies have demonstrated that Tau22 mice We next examined the phosphorylation level of 17 exhibit impaired long-term depression (LTD) but nor- kinases and signaling molecules by using the MAPK mal long-term potentiation (LTP) at the Schaffer collat - Phosphorylation Array (RayBiotech, GA, USA). Only eral-CA1 synapses in the hippocampus [33]. As shown minor or no change was found between Tau22 mice and in Fig.  1g, LTD was maintained in WTC mice but not in WT mice (Fig. S4). No marked alterations in the level or TauC mice. This impairment in LTD was prevented by activation of PP2A [46], the major tau phosphatase, were J4. The average LTD magnitude during the last 10  min detected either (Additional file 1: Fig. S5). of recording was quantified and shown in Fig.  1h. Col- Because adenosine homeostasis has been implicated in lectively, J4 normalizes the impaired basal synaptic trans- the regulation of AMPK activation [31, 32, 47] and since mission and LTD at Schaffer collateral synapses in Tau22 AMPK directly phosphorylates Tau [48], we evaluated hippocampi without affecting those in WT hippocampi. the activation/phosphorylation of AMPK (Thr , des- ignated pAMPK) and Tau phosphorylation in the hip- Chronic J4 treatment reduces Tau hyperphosphorylation pocampi of postmortem Alzheimer’s Disease, FTLD-Tau in the hippocampi of Tau22 mice patients, and Tau22 mice. Immunofluorescence staining To examine the effect of J4 on Tau phosphorylation [33], revealed that, in the posterior hippocampal sections from we performed a differential peptide  labeling  (iTRAQ) Alzheimer’s Disease and FTD-Tau patients, pAMPK was of hippocampal proteins and analyzed the results using detected in neurons that contained phosphorylated Tau LC–MS/MS-based proteomics. Phosphopeptides cover- (Figs.  3a and Additional file  1: S6). Similarly, elevated ing 30 phosphorylation sites of human Tau were iden- AMPK phosphorylation was observed in neurons con- tified (Fig.  2a). In total, J4 reduced phosphorylation at taining phosphorylated Tau in the CA1 region in Tau22 15 phosphorylation sites. The J4-mediated reduction mice, but rarely in those in WT mice (Fig. 3b, c). J4 treat- in Tau phosphorylation at sites commonly observed in ment decreased the levels of pAMPK and pTau in Tau22 (See figure on next page.) Fig. 1 Chronic treatment with J4 alleviates the impairment of spatial memory and hippocampal CA1 LTD in Tau22 mice. a–d Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green; n = 12–20 in each group) from the age of 3–10 months. The Morris water maze (MWM) with a hidden platform was used to assess spatial learning and memory. a The acquisition‑learning phase (Day 1–Day 5) and the reversal‑learning phase (Day 9–Day 12) of MWM. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group; two‑ way ANOVA. Probe tests for b the spatial memory and c spatial reversal memory were performed on Day 8 and Day 15, respectively. The average percentage of time spent in the T (target quadrant) and NTs (nontarget quadrants) were calculated. *p < 0.05 versus the NTs; two‑tailed Student’s t-test. d The swimming speed (cm/s) of the animals in the probe test. e–h Hippocampal slices were prepared from mice subjected to different treatment groups (n = 3–18 slices from 3 to 8 mice in each group) from the age of 3–10 months. e The input/output relationship curve (fEPSP responses, stimuli strengths increased from 10 to 120 μA), f averaged paired‑pulse ratios (interstimulus intervals, from 50 to 500 ms), and g, h Long‑term depression (LTD) induced by 3 trains of LFS (2 Hz, 1200 pulses) at the Schaffer collateral‑ CA1 synapses were recorded. g The average fEPSP slopes of mice subjected to different treatments were calculated. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group, two‑ way ANOVA. h Quantification results of the mean fEPSP slopes during the last 10 min of the steady‑state period. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group, one‑ way ANOVA. The data are expressed as means ± S.E.M Chang  et al. acta neuropathol commun (2021) 9:112 Page 7 of 18 Chang et al. acta neuropathol commun (2021) 9:112 Page 8 of 18 mice compared to WT mice (Fig. 3b–d). In line with this Tauopathy-associated genes. Gene ontology (GO; Fig. 4c, finding, analyses of hippocampal proteins using a com - FDR < 0.05, Benjamini p < 0.01) and Kyoto Encyclope- bination of iTRAQ and LC–MS/MS-based proteomics dia of Genes and Genomes (KEGG; Fig.  4d, FDR < 0.05, by Phosphopeptides revealed that the phosphorylation Benjamini p < 0.01) analyses were conducted, and the levels of two AMPK downstream targets (i.e., eukary- DE genes between the TauC group vs WTC group were otic elongation factor 2 (eEF2) and myosin VI (Myo6); found to be associated with the immune response and 2+ [49, 50]) and one upstream regulator (Ca /calmodulin- transcriptional machinery. Most of the pathways dereg- dependent protein kinase kinase-β, Camk2b; [51]) were ulated by Tau pathology (red bars in Fig.  4c, d) became elevated in Tau22 mice. J4 treatment reduced the phos- less significant after J4 treatment (green bars). When we phorylation of eEF2, Myo6, and Camk2b (Additional analyzed the upregulation- or downregulation-specific file  1: Table S6), supporting that J4 normalized the upreg- DE genes separately (Additional file  1: Fig. S7A–S7C and ulated AMPK signaling pathway in the hippocampus of S7D–S7F, respectively), multiple pathways (e.g., tran- Tau22 mice. scription-related machineries, angiogenesis, cell–cell Since AMPK is a critical energy sensor and Tau has interaction, cell adhesion) remained markedly normal- been implicated in mitochondrial dysfunction [52], we ized by J4 treatment. Part of the inflammation-related next assessed mitochondrial mass by immunohisto- pathways were also rescued by J4 treatment. chemical staining using an antibody against ATP5a, a Next, we specifically analyzed the effect of J4 on Tau22 component of complex V [53]. Consistent with abnormal mice vs. control Tau22 mice. A total of 1239 DE genes AMPK activation, TauC mice exhibited less ATP5a-posi- were identified (289 upregulated and 950 downregulated; tive mitochondrial mass in the hippocampus than WTC TauJ mice versus TauC mice, p < 0.05 and absolute log mice (Fig.  3e). J4 reversed mitochondrial loss in Tau22 fold-change ≥ 0.32; Fig.  4e). Importantly, J4 normalized mice (Fig.  3e, f ) as it normalized the AMPK overactiva- the expression of 436 of the 950 upregulated DE genes tion (Fig.  3b, c), suggesting that the blockade of ENT1 (45.8%; Fig.  4f) and 85 of the 491 downregulated DE normalized Tau-associated energy dysfunction. genes (17.3%, data not shown) between the TauC group and WTC group. Moreover, J4 normalized multiple dys- Transcriptomic signature associated with the beneficial regulated canonical pathways, which were also observed effect of J4 in the hippocampi of Tau22 mice in other Alzheimer’s disease mouse models (Tg4510 To gain mechanistic insight into J4’s action, we per- and APP/PS1), in Tau22 mice (Additional file  1: Fig. formed transcriptional profiling of the hippocampus S8). Overall, J4 had a broad impact on Alzheimer’s Dis- using RNA-seq analysis. A total of 1441 differentially ease-related signaling molecules and pathways in Tau22 expressed (DE) genes (950 upregulated and 491 down- hippocampi. regulated) between TauC and WTC mice were identified (Fig. 4a, TauC mice versus WTC mice, p < 0.05 and abso- Chronic J4 treatment mitigates the activation of microglia lute log fold-change ≥ 0.32). Between the TauJ and WTC in the hippocampi of Tau22 mice groups, a total of 1304 DE genes were also identified Based on the cell-type information listed in the Brain (Fig.  4b; TauJ mice versus WTC mice, p < 0.05, absolute RNAseq database (https:// www. brain rnaseq. org, [54]), log fold-change ≥ 0.32) but the proportion of upregu- we further classified the 436 DE genes with expres - lated (417) and downregulated (887) markedly differed. sion levels normalized by J4 (Fig.  4f) into five types Interestingly, only a fraction (216 upregulated; 214 down- (including neuron-enriched, glia-enriched, endothelial regulated; (Additional file  1: Fig. S7A and S7D) of the cell-enriched, and unclassified). As shown in Fig.  4g, genes affected in Tau mice as compared to WT, remained approximately 55% of the DE genes whose expression lev- altered in the Tau mice after J4 treatment. These data sug - els that were normalized by J4 were enriched in glial cells, gested that J4 significantly normalized the expressions of including astrocytes and microglia (Fig.  4g). In Tau22 (See figure on next page.) Fig. 2 Chronic J4 treatment decreases hyperphosphorylated and misfolded human Tau levels in the hippocampi of Tau22 mice. Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green) for 8–9 months from the age of 3 months. a Pooled total hippocampal lysates (200 μg) from 3 animals of the age of 12 months were harvested and subjected to phospho‑proteomic analysis. The heatmap shows the relative log2 expression ratio of phosphorylated human tau (MAPT, P10636) in the TauJ group vs. the TauC group. The relative expression level (log2 ratio) of human phosphorylated tau is shown on a scale from red (upregulated) to blue (downregulated). The identified phosphorylation sites on peptides are shown in bold and underlined. b, c Hippocampal sections (20 μm) were prepared from mice with different treatment groups (n = 3–5 in each group) from the age of 3–11 months and subjected to IHC staining. b The levels of hyperphosphorylated tau and misfolded tau in the hippocampus were evaluated by staining with the indicated antibodies (AT8 for Ser202/Thr205, green; AT100 for Thr212/Ser214, green; MC1 for conformational changed tau, green), and the quantification results are shown in (c). Scale bar, 50 μm. The data are expressed as the mean ± S.E.M. p < 0.05, versus the TauC group, two‑tailed Student’s t-test Chang  et al. acta neuropathol commun (2021) 9:112 Page 9 of 18 C Chang et al. acta neuropathol commun (2021) 9:112 Page 10 of 18 Fig. 3 Chronic J4 treatment decreases AMPK activation and rescues mitochondrial abnormalities in the hippocampi of Tau22 mice. a Posterior hippocampal sections (6 μm) from normal subjects and Alzheimer’s Disease and FTD‑ Tau (CBD, PSP, and Pick’s disease) patients were subjected to Thr172 IHC staining. The levels of phospho‑AMPK and hyperphosphorylated tau were evaluated by staining with the indicated antibodies (pAMPK , Ser202/Thr205 green; AT8 for pTau , red). b–f Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green; n = 5–7 in each group) from Thr172 the age of 3–11 months. Hippocampal sections (20 μm) were prepared and subjected to IHC staining using the indicated antibodies (pA MPK , Ser202/Thr205 green; AT8 for pTau , red), and the staining was quantified (c, d). Scale bar, 20 μm. e, f The level of the mitochondrial marker ATP5a was evaluated by staining with an anti‑ATP5a antibody (e, green) and quantified (f; n = 3 in each group). Scale bar, 5 μm. The data are expressed as the mean ± S.E.M. *p < 0.05 versus the WTC group; p < 0.05, versus the TauC group, one‑ way ANOVA hippocampi, enhanced gene signatures for the disease- mice (Fig.  5a, b). We also measured the transcript lev- associated microglia were detected (DAM, [55]; Fig.  4h els of several factors secreted by activated microglia and and Additional file  1: Table S7). Consistently, the immu- known to favor neurotoxic activation of astrocytes (A1 noreactive intensities of both Iba1 (a marker of micro- phenotype; [56]) using RT-qPCR. As shown in Fig.  5c, glia) and CD68 (a marker of activated microglia) were the levels of TNF-α and C1q (C1qa, C1qb, and C1qc), but significantly elevated in the hippocampi of TauC mice not IL-1α, were upregulated in the Tau22 hippocampi (11 months old) compared with the hippocampi of WTC compared with WT hippocampi. Immunofluorescence Chang  et al. acta neuropathol commun (2021) 9:112 Page 11 of 18 staining and RT-qPCR showed that J4 prevented the secondary to microglial activation in Tau22 hippocampi upregulation of TNF-α, C1q, and CD68 in Tau22 mice since activated microglia (CD68-positive) were detected (Fig.  5a–e, Additional file  1: S9A and S9B), suggesting in the hippocampi of young Tau22 mice (4  months that J4 mitigated microglial inflammation in Tau22 mice. old), when no reactive astrocytes (Lcn2-positive) were observed (Additional file 1: Fig. S11). Chronic J4 treatment mitigates synaptic loss in Tau22 mice According to the reduction of the microglial phenotype C1q is an important mediator of Tau-induced synaptic as well as TNFα and C1q expressions (Fig. 5c), chronic J4 loss [34, 57]. We thus examined the number of synapses treatment normalized not only GFAP and Lcn2 levels but by immunofluorescence staining. In Tau22 hippocampi, also the pathological upregulation of A1-specific genes the levels of a postsynaptic marker (PSD95) and a pre- expression (Figs.  4i, 5h, i; Additional file  1: Table  S8). synaptic marker (synaptophysin) were lower than those Collectively, our data suggest that early Tau-induced in WT hippocampi. Consistent with the rescuing effect microglial activation is likely to promote the activation of of J4 on C1q, J4 restored the levels of PSD95 and synap- neurotoxic astrocytes and can be blocked by J4. tophysin in Tau22 mice (Additional file  1: Fig. S9C and S9D). We also determined the number of synapses based Discussion on the colocalization of PSD95 and synaptophysin. In The present study showed that chronic treatment with J4, line with the reduction in CD68 and C1q levels (Fig. 5a– an ENT1 blocker, mitigates Tau pathology by alleviating c), J4 rescued the synaptic loss in Tau22 hippocampi [58] not only mitochondrial dysfunction and AMPK overac- (Fig. 5f, g). tivation but also the neuroinflammatory status of micro - glia and astroglia, ultimately attenuating the impairment Chronic J4 treatment suppresses the cytotoxic astrocytes of compromised synapses as well as spatial learning and induction in the hippocampi of Tau22 mice memory. Our study particularly supports a functional TNF-α and C1q are potent astrocytic activators for the link between adenosine homeostasis, AMPK regulation neurotoxic A1 phenotype [56]. Our RNA-seq analysis of and Tau pathology development. Tau22 hippocampi revealed the upregulation of gene sig- Mitochondrial dysfunction is a major pathogenic fea- natures of pan-reactive and cytotoxic A1 astrocytes ([59], ture of Alzheimer’s Disease [61] and is known to facili- Additional file  1: Table  S8 and Fig.  4i). Immunofluores - tate the hyperphosphorylation of Tau, which in turn cence staining further showed that Tau22 mice (TauC) alters the morphology and functions of mitochondria exhibited higher levels of GFAP (an astrocyte marker) [62]. Therefore, it is not surprising that AMPK, a key and Lcn2 (a pan reactive astrocyte marker, [60]) in their energy sensor and an upstream kinase of Tau, is over- hippocampus than WT mice (WTC; Fig. 5h, i), confirm - activated in the hippocampi of patients with Alzhei- ing that astrocytes in Tau22 hippocampi were abnormally mer’s Disease or tauopathies [28]. One major function activated. Interestingly, FTD-Tau patients and late stage of AMPK is the maintenance of cellular energy homeo- Alzheimer’s Disease patients (Additional file  1: Fig. S10), stasis through modulation of the balance between ana- upregulation of several A1-specific genes (e.g., GBP2, bolic and catabolic processes [29]. Because hippocampal SERPING1, FKBP5) was observed. Notably, the five A1 neurons of WT mice are homeostatic in nature, no sig- astrocyte genes tested were all significantly upregulated nificant AMPK activation was observed in the hippocam - in the frontal cortex of FTD-Tau-Pick’s disease patients pus of WT mice (Fig.  3b, c). Treatment with J4 showed compared to control subjects, suggesting a link between no impact on AMPK activation in such a homeostatic Tau and astrocyte reactivity (Additional file  1: Fig. condition, suggesting that ENT1 does not play a sig- S10). The induction of reactive A1 astrocytes appears nificant role in the regulation of AMPK in physiological (See figure on next page.) Fig. 4 Chronic J4 treatment ameliorates the expression of tauopathy associated genes and neurotoxic reactive astrocyte genes in the hippocampi of Tau22 mice. Mice were treated as indicated ( WTC, TauC, and TauJ; n = 3 in each group) from the age of 3–10 months. The hippocampus was carefully removed for RNA‑seq analysis. Volcano plot of DE genes in the a TauC group vs. WTC group and b TauJ group vs. WTC group. c GO enrichment analysis of DE genes between the TauC and WTC groups (red bar) and TauJ and WTC groups (green bar). d KEGG pathway analysis of DE genes between the TauC and WTC groups (red bar) and TauJ and WTC groups (green bar). e Volcano plot of the DE genes in the TauJ group vs. TauC group. In the volcano plot, the significantly upregulated and downregulated DE genes (absolute log ratio ≥ 0.32; p < 0.05) are shown in red and blue, respectively. f Venn diagram showing the 436 overlapping DE genes between the upregulated DE genes in the TauC group vs. WTC group (pink) and the downregulated DE genes in the TauJ group vs. TauC group (green). g Pie chart of cell‑type ‑ enrichment of 436 tauopathy‑associated DE genes regulated by J4. In the volcano plot, the significantly upregulated and downregulated DE genes (absolute log ratio ≥ 0.32; p < 0.05) are shown in red and blue, respectively. In the Venn diagram and pie chart, the number and percentage of DE genes in each category were shown in each sector. h, i Heatmap of h DAM genes and i A1‑specific genes in the TauC/WTC and TauJ/TauC groups. The relative expression level (log ratio) of genes is shown on a scale from red (upregulated) to blue (downregulated). Asterisks indicate significant alterations (p < 0.05) Chang et al. acta neuropathol commun (2021) 9:112 Page 12 of 18 Chang  et al. acta neuropathol commun (2021) 9:112 Page 13 of 18 conditions. Conversely, the impaired energy status of microenvironments (e.g., extracellular space proximal to hippocampal neurons of Tau22 mice (i.e., in an allostatic neuronal soma and synapses), future investigations using situation) causes the activation of AMPK. We hypoth- an in  vivo adenosine sensor [64] will be needed. None- esized that blockade of ENT1 may reduce the entry of theless, the levels of at least three genes (i.e., ADA, CD39, adenosine and, subsequently the cellular level of AMP, CD73) involved in adenosine metabolism were elevated thereby altering the AMP/ATP ratio, and ultimately sup- in the hippocampus of Tau22 mice. A trend of increase pressing AMPK activation in hippocampal neurons of in the transcript level of ADK was also found, but did not Tau22 mice. Our hypothesis is in line with a recent study reach statistical significance. Importantly, J4 treatment demonstrating that genetic deletion of ENT1 in erythro- normalized all these changes (Table  1), suggesting that cytes reduces adenosine uptake and leads to the suppres- adenosine homeostasis was altered in Tau22 mice. It is sion of AMPK [47]. possible that the elevation of ADA in the hippocampus of Accumulating evidence demonstrates that overacti- Tau22 mice may reduce adenosine availability from intra- vation of AMPK in neurons causes synapse loss via an cellular source, while the elevation of CD39 and CD73 autophagy-dependent pathway, and links synaptic integ- increases extracellular adenosine pool, which counteracts rity and energetic failure in neurodegenerative diseases the imbalance of intracellular adenosine level. This may [63]. Here, we found that aberrant AMPK activation was be why the adenosine alteration was not observed in the associated with synaptic loss and reduced basal synaptic hippocampus of Tau22 mice (Additional file  1: Fig. S14). transmission in Tau22 hippocampi (Figs.  1e, 3b, c, 5f, g; J4 treatment reset adenosine homeostasis by blockading Additional file  1: Table  S6). Collectively, J4 suppresses adenosine entry, which results in the decrease of ADA, AMPK overactivation, and normalizes impaired neuronal CD39, and CD73 in the hippocampus of Tau22 mice. plasticity in both APP/PS1 and Tau22 mice (LTP and Collectively, chronic J4 does not induce a major change LTD, respectively; [27]; Fig. 1g, h). in the steady state level of adenosine but rather adenosine Besides the impaired cognitive function, we did not homeostasis. observe any obvious systemic alteration of Tau22 up to An interesting study recently reported that Tau22 mice 12 months except for a slightly lower body weight. Treat- are more susceptible to pentylenetetrazol (PTZ) for sei- ment with J4 did not affect the bodyweight of Tau22 and zure and mortality than WT mice, probably due to the WT mice (Additional file  1: Fig. S12), suggesting that enhanced expression of ADK [65]. This is of great inter - chronic J4 treatment at the condition tested had no obvi- est because J4 is an anti-epileptic agent in a PTZ-induced ous toxicity. Although significant Tau hyperphosphoryla - kindling model [66]. Given that J4 treatment reduced tion and gliosis were observed in the hippocampus of the level of ADK in Tau22 mice (Table  1), it is plausible Tau22 mice of 10–12  months (Figs.  2 and 5), no change that modulation of ADK may contribute to the beneficial in the volume of the whole brain, hippocampus, and ven- effect of J4 in Tau22 mice. No effect of J4 on the expres - tricle were altered (Additional file  1: Fig. S13). Because J4 sions of A adenosine receptor, A adenosine receptor 1 2A is a blocker of ENT1, we measured the levels of adeno- and some of their signaling molecules (i.e., protein kinase sine in the hippocampus of Tau22 mice (11 months old) A, GSK3β, AMPKs) were observed (Additional file  1: but found no difference in either the extracellular or Table S9). the intracellular steady state levels by in  vivo microdi- Although pathogenic Tau is specifically expressed in alysis and tissue extraction coupled to high performance the neurons of mouse models of tauopathy (including liquid chromatography (HPLC), respectively (Figs. S1 Tau22 and rTg4510 mice), reactive microglia and astro- and S14). This is probably because what we measured cytes have been found near neurons that contain high were bulk adenosine concentrations in the extracellu- levels of pathogenic Tau, suggesting that degenerating lar fluid and those inside of cells in the hippocampus. neurons may trigger abnormal gliosis [33] and alter their To assess whether adenosine levels are altered in gene expression profiles [67]. Here, our RNA-seq analysis (See figure on next page.) Fig. 5 Chronic J4 treatment suppresses the activation of microglia and the induction of cytotoxic A1 astrocytes in the hippocampi of Tau22 mice. Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green; n = 5–7 in each group) from the age of 3–11 months, and their tissues were subjected to IHC staining and RT‑ qPCR analysis. a CD68 (green) is a marker of reactive microglia and Iba1 (red) is a marker of microglia and the quantification results are shown in (b). c Gene expression of TNF‑α, IL ‑1α, and C1q (C1qa, C1qb, and C1qc) in the hippocampi of treated mice (n = 6–9 in each group) was analyzed, and GAPDH was used as a reference gene. d The intensity of C1q (green) and Iba1 (red) expression was examined and the quantification results are shown in (e). Scale bar, 20 μm. f, g The number of synapses (f) in the hippocampi of treated mice (n = 10–12 in each group) was evaluated by staining with the indicated antibodies (PSD95, green; SYP, red). The quantification results are shown in (g). Scale bar, 5 μm. h Hippocampal sections (20 μm) were prepared and subjected to IHC staining (Lcn2 a marker of reactive astrocytes, green; GFAP a marker of astrocytes, red) and the staining was quantified (i). Scale bar, 20 μm. The data are expressed as the mean ± S.E.M. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group, one‑ way ANOVA Chang et al. acta neuropathol commun (2021) 9:112 Page 14 of 18 of Tau22 mice showed upregulation of several genes not rescue all inflammation-related pathways as analyzed associated with the disease-associated microglia (DAM, by GO and KEGG (Additional file  1: Fig. S7), 56 of the [55]) in mice and patients with Alzheimer’s Disease 90 dysregulated homeostatic microglia genes (62%; [68]) (Fig.  4h and Additional file  1: Table S7). Although J4 did of Tau22 mice were normalized by J4 treatment (Fig. Chang  et al. acta neuropathol commun (2021) 9:112 Page 15 of 18 S15). These homeostatic microglial genes are commonly Conclusions expressed in microglia of healthy adult brain. Thus, J4 In summary, we provide evidence that an ENT1 inhibi- treatment rescued the dysregulated microglial homeosta- tor (J4) rescues the energy dysfunction (including mito- sis in Tauopathy. chondrial impairment and AMPK overactivation) and Notably, J4 not only ameliorated the energy dysfunc- pathological glial activation, and subsequently improves tion and Tau pathology in neurons (Figs.  1, 2, 3) but synaptic function and memory in tauopathy. Modula- also markedly reduced the activation of CD68-positive tion of adenosine homeostasis by an ENT1 inhibitor (J4) microglia activation and rescued the synapse loss asso- therefore deserves further development in tauopathies ciated with the phagocytic activities of activated micro- and Alzheimer’s Disease. glia (Fig. 5) [34, 57, 69]. Given that CD68 is also a marker of phagocytosis [70], the suppression of CD68 by J4 sug- Abbreviations gest that J4 treatment might decrease the phagocytic Aβ: Amyloid‑beta; ACN: Acetonitrile; ACSF: Artificial cerebrospinal fluid; AD: capacity of microglia and result in the rescue of synapse Alzheimer’s disease; ADA: Adenosine deaminase; ADK: Adenosine kinase; AMPK: AMP‑activated protein kinase; BBB: Blood–brain barrier; BSA: Bovine loss (Fig.  5b), in agreement with the normalization of serum albumin; CBD: Corticobasal degeneration; cDNA: Complementary C1q expression by J4, a microglial complement protein DNA; DAAs: Disease‑associated astrocytes; DAM: Disease ‑associated microglia; known to regulate synaptic phagocytosis by microglia DAVID: Database for Annotation, Visualization and Integrated Discovery; DE: Differentially expressed; DTT: Dithiothreitol; ENT1: Equilibrative nucleoside in a Tau pathology context [34, 69]. We also performed transporter 1; FDR: False discovery rate; fEPSPs: Field excitatory postsynaptic RT-qPCR and established that microglial factors known potentials; FPKM: Fragments per kilobase per million; FTD: Frontotemporal to promote neurotoxic activation of astrocytes were dementia; GO: Gene Ontology; HPβCD: 2‑Hydroxypropyl‑b ‑ cyclodextrin; IAA: Iodoacetamide; IACUC : Institutional Animal Care and Utilization Committee; upregulated in Tau22 mice (C1q and TNFα; Fig.  5c). IVC: In ventilated cages; KEGG: Kyoto Encyclopedia of Genes and Genomes; Consistent with the suppression of microglial activa- LFS: Low‑frequency stimulation; LTD: Long‑term depression; LTP: Long‑term tion, J4 also significantly reduced astrocytic activation, potentiation; MWM: Morris water maze; NDS: Normal donkey serum; NGS: Normal goat serum; PCF: Proteomics Core Facility; PP2A: Protein phosphatase with a particular impact on the A1 signature (Figs.  4i, 2A; PPF: Paired‑pulse facilitation; PSP: Progressive supranuclear palsy; RT ‑ qPCR: 5h, i). We show, for the first time, a pathological acti - Reverse Transcription quantitative PCR; PTZ: Pentylenetetrazol; SBS: Sequenc‑ vation of the A1 neurotoxic astrocytic phenotype in a ing‑by‑synthesis; Tau22: THY ‑ Tau22; TauC: Vehicle‑treated Tau22 mice; TauJ: J4‑treated Tau22; TEAB: Triethylammonium bicarbonate; TFA: Trifluoroacetic tauopathy context using both human and mouse sam- acid; WT: Wild type; WTC : Vehicle‑treated WT mice; WTJ: J4‑treated WT. ples (Additional file  1: Fig. S10 and Table  S8), which may be related to the production of A1-promoting fac- Supplementary Information tors by microglia. Our RNA-seq analysis of Tau22 mice The online version contains supplementary material available at https:// doi. also emphasizes the upregulation of a disease-associ- org/ 10. 1186/ s40478‑ 021‑ 01213‑7. ated astrocytes (DAAs) signature, recently detected in an amyloid Alzheimer’s Disease mouse model (5XFAD) Additional file 1. Supplementary Information. [71] (Additional file  1: Table  S10). These observations are of particular importance since glial activation has Acknowledgements been shown to be instrumental in Alzheimer’s Disease We are grateful to the Proteomics Core Facility of the Institute of Biomedical [5]. In accordance, the beneficial effects of J4 on mem - Sciences, Academia Sinica, and the Neuroscience core facility, Academia Sinica for the LC/MS/MS analysis and the electrophysiology analysis. We thank the ory alterations and neuroplasticity (Figs. 1, 2, 3) is asso- Animal Image Facility, Academia Sinica and Taiwan Animal Consortium for the ciated with the normalization of both A1-promoting technical support in MRI analysis. We acknowledge the UC Davis Alzheimer’s factors and A1 astrocytes in Tau22 mice (Additional Disease Center (USA) for providing human brain specimens from normal, Alzheimer’s Disease, and FTD‑ Tau subjects. We also thank Dr. Peter Davies ( The file  1: Table  S8). Treatment with J4 normalized 60% of Feinstein Institute for Medical Research, New York, USA) for generously provid‑ the elevated DAA genes in Tau22 mice (Additional ing disease‑specific conformational modified tau antibody (MC1). file  1: Table S10), further supporting the beneficial effect Authors’ contributions of J4 on the prevention of abnormal astrocytic activa- CPC conducted behavioral, bioinformatic analysis, and immunostaining of tion. Noteworthy, ENT1 is not only expressed by neu- human brain specimens. YGC performed electrophysiology experiments. rons but also by astrocytes. Earlier studies had reported PYC and TNAN performed biochemical analyses. KCW and FYC performed pharmacokinetics studies. HMC and TNAN managed the animal breeding and that blockade of ENT1 in astrocytes suppresses the genotyping. LWJ and VH provided human brain specimens. KC performed expression of two astrocyte-specific genes (i.e., the Q‑PCR analysis of human brain specimens. LB and DB provided THY ‑ Tau22 type 2 excitatory amino acid transporter (EAAT2) and mice and contributes to bioinformatic analysis. YC, YFL, CJL, and SJC super‑ vised the design of biochemical analyses, electrophysiology experiments, and aquaporin 4 (AQP4)) [72]. Because J4 treatment did not pharmacokinetics studies. CPC, SJC, KC and DB contributed to manuscript affect the level of EAAT2 and AQP4 (Additional file  1: writing. YC oversaw project administration and wrote the final manuscript. All Table  S9), suppression of astrocytic ENT1 by J4 under authors read and approved the final manuscript. the condition tested was unlikely to be significant. Chang et al. acta neuropathol commun (2021) 9:112 Page 16 of 18 Funding 6. Qian W, Shi J, Yin X, Iqbal K, Grundke‑Iqbal I, Gong CX, Liu F (2010) PP2A This research was supported by the Academia Sinica and Ministry of Science regulates tau phosphorylation directly and also indirectly via activat‑ and Technology (MOST 107‑2320‑B‑001‑013‑MY3, AS‑SUMMIT ‑109; MOST ‑ ing GSK‑3beta. J Alzheimers Dis 19:1221–1229. https:// doi. org/ 10. 3233/ 108‑3114‑ Y‑001‑002; AS‑KPQ ‑109‑BioMed). The Neuroscience core facility was jad‑ 2010‑ 1317 supported by the Academia Sinica (AS‑ CFII‑108‑106). The UC Davis Alzheimer’s 7. Sergeant N, Bretteville A, Hamdane M, Caillet‑Boudin ML, Grognet P, Disease Center Biorepository (ADC Biorepository) was supported by the Bombois S, Blum D, Delacourte A, Pasquier F, Vanmechelen E et al (2008) National Institutes of Health (P30‑AG010129). The “Alzheimer & Tauopathies” Biochemistry of Tau in Alzheimer’s disease and related neurological dis‑ laboratory is supported by Inserm, Université Lille, France Alzheimer, programs orders. Expert Rev Proteomics 5:207–224. https:// doi. org/ 10. 1586/ 14789 d’investissements d’avenir LabEx (excellence laboratory) DISTALZ (Develop‑450.5. 2. 207 ment of Innovative Strategies for a Transdisciplinary approach to ALZheimer’s 8. Lee CW, Shih YH, Wu SY, Yang T, Lin C, Kuo YM (2013) Hypoglycemia disease), ANR (ADORASTrAU ANR‑18‑ CE16‑0008 and CoEN 5008), Fondation induces tau hyperphosphorylation. Curr Alzheimer Res 10:298–308. pour la Recherche Médicale, Vaincre Alzheimer, Fondation Plan Alzheimer https:// doi. org/ 10. 2174/ 15672 05011 31003 0009 (ADOMEMOTAU), LilleMétropole Communauté Urbaine, Région Hauts‑ de‑ 9. Merkwirth C, Martinelli P, Korwitz A, Morbin M, Bronneke HS, Jordan SD, France (COGNADORA), and DN2M. Special thanks to PHC Orchid exchange Rugarli EI, Langer T (2012) Loss of prohibitin membrane scaffolds impairs funding for supporting the collaboration between Academia Sinica and mitochondrial architecture and leads to tau hyperphosphorylation and Inserm. neurodegeneration. PLoS Genet 8:e1003021. https:// doi. org/ 10. 1371/ journ al. pgen. 10030 21 Availability of data and materials 10. Carman AJ, Mills JH, Krenz A, Kim DG, Bynoe MS (2011) Adenosine The data that support the findings of this study are available on request to the receptor signaling modulates permeability of the blood–brain barrier. J corresponding author. Neurosci 31:13272–13280. https:// doi. org/ 10. 1523/ JNEUR OSCI. 3337‑ 11. Ethical approval and consent to participate 11. Chen JF, Lee CF, Chern Y (2014) Adenosine receptor neurobiology: All animal studies were conducted following the protocols approved by the overview. Int Rev Neurobiol 119:1–49. https:// doi. org/ 10. 1016/ B978‑0‑ 12‑ Institutional Animal Care and Utilization Committee (IACUC, Academia Sinica, 801022‑ 8. 00001‑5 Taiwan). Human brain samples were obtained from the UC Davis Alzheimer’s 12. Ham J, Evans BA (2012) An emerging role for adenosine and its receptors Disease Center (USA) and the brain banks of Lille, Paris, and Geneva. The IRB in bone homeostasis. Front Endocrinol (Lausanne) 3:113. https:// doi. org/ number of Academic Sinica is AS‑IRB‑BM ‑16013.10. 3389/ fendo. 2012. 00113 13. Lee CF, Chern Y (2014) Adenosine receptors and Huntington’s disease. Int Consent for publication Rev Neurobiol 119:195–232. https:// doi. org/ 10. 1016/ B978‑0‑ 12‑ 801022‑ 8. Not applicable. 00010‑6 14. Anderson CM, Xiong W, Geiger JD, Young JD, Cass CE, Baldwin SA, Parkin‑ Competing interests son FE (1999) Distribution of equilibrative, nitrobenzylthioinosine‑sen‑ Yijuang Chern hold patents on J4 for the treatment of neurodegenerative sitive nucleoside transporters (ENT1) in brain. J Neurochem 73:867–873. diseases.https:// doi. org/ 10. 1046/j. 1471‑ 4159. 1999. 07308 67.x 15. Young JD (2016) The SLC28 (CNT ) and SLC29 (ENT ) nucleoside trans‑ Author details porter families: a 30‑ year collaborative odyssey. Biochem Soc Trans Institute of Biomedical Sciences, Academia Sinica, Nankang, Taipei 115, 44:869–876. https:// doi. org/ 10. 1042/ BST20 160038 Taiwan. School of Pharmacy, National Taiwan University, Taipei, Taiwan. 16. Cunha RA (2016) How does adenosine control neuronal dysfunction and Neuroscience Program of Academia Sinica, Academia Sinica, Taipei, Taiwan. neurodegeneration? J Neurochem 139:1019–1055. https:// doi. org/ 10. Department of Pathology and Laboratory Medicine, University of California 1111/ jnc. 13724 Davis, Sacramento, CA, USA. Univ. Lille, Inserm, CHU Lille, U1172 ‑ LilNCog ‑ 17. Porkka‑Heiskanen T, Kalinchuk AV (2011) Adenosine, energy metabolism Lille Neuroscience & Cognition, 59000 Lille, France. Alzheimer & T auopathies, and sleep homeostasis. Sleep Med Rev 15:123–135. https:// doi. org/ 10. LabEx DISTALZ, LiCEND, 59000 Lille, France. Institute of Cellular and Organis‑1016/j. smrv. 2010. 06. 005 mic Biology, Academia Sinica, Taipei, Taiwan. 18. Alonso‑Andres P, Albasanz JL, Ferrer I, Martin M (2018) Purine ‑related metabolites and their converting enzymes are altered in frontal, parietal Received: 21 April 2021 Accepted: 7 June 2021 and temporal cortex at early stages of Alzheimer’s disease pathology. Brain Pathol 28:933–946. https:// doi. org/ 10. 1111/ bpa. 12592 19. Albasanz JL, Perez S, Barrachina M, Ferrer I, Martin M (2008) Up‑regulation of adenosine receptors in the frontal cortex in Alzheimer’s disease. Brain Pathol 18:211–219. https:// doi. org/ 10. 1111/j. 1750‑ 3639. 2007. 00112.x 20. Temido‑Ferreira M, Ferreira DG, Batalha VL, Marques‑Morgado I, Coelho References JE, Pereira P, Gomes R, Pinto A, Carvalho S, Canas PM et al (2018) Age‑ 1. Cummings J, Blennow K, Johnson K, Keeley M, Bateman RJ, Molinuevo JL, related shift in LTD is dependent on neuronal adenosine A2A receptors Touchon J, Aisen P, Vellas B (2019) Anti‑tau trials for Alzheimer’s disease: a interplay with mGluR5 and NMDA receptors. Mol Psychiatry. https:// doi. report from the EU/US/CTAD task force. J Prev Alzheimers Dis 6:157–163. org/ 10. 1038/ s41380‑ 018‑ 0110‑9 https:// doi. org/ 10. 14283/ jpad. 2019. 14 21. Faivre E, Coelho JE, Zornbach K, Malik E, Baqi Y, Schneider M, Cellai L, 2. Brier MR, Gordon B, Friedrichsen K, McCarthy J, Stern A, Christensen J, Carvalho K, Sebda S, Figeac M et al (2018) Beneficial effect of a selective Owen C, Aldea P, Su Y, Hassenstab J et al (2016) Tau and Abeta imaging, adenosine A2A receptor antagonist in the APPswe/PS1dE9 mouse model CSF measures, and cognition in Alzheimer’s disease. Sci Transl Med of Alzheimer’s disease. Front Mol Neurosci 11:235. https:// doi. org/ 10. 8:338ra366. https:// doi. org/ 10. 1126/ scitr anslm ed. aaf23 62 3389/ fnmol. 2018. 00235 3. Colin M, Dujardin S, Schraen‑Maschke S, Meno ‑ Tetang G, Duyckaerts C, 22. Lee CC, Chang CP, Lin CJ, Lai HL, Kao YH, Cheng SJ, Chen HM, Liao YP, Courade JP, Buee L (2020) From the prion‑like propagation hypothesis Faivre E, Buee L et al (2018) Adenosine augmentation evoked by an ENT1 to therapeutic strategies of anti‑tau immunotherapy. Acta Neuropathol inhibitor improves memory impairment and neuronal plasticity in the 139:3–25. https:// doi. org/ 10. 1007/ s00401‑ 019‑ 02087‑9 APP/PS1 mouse model of Alzheimer’s disease. Mol Neurobiol 55:8936– 4. Ittner LM, Gotz J (2011) Amyloid‑beta and tau—a toxic pas de deux in 8952. https:// doi. org/ 10. 1007/ s12035‑ 018‑ 1030‑z Alzheimer’s disease. Nat Rev Neurosci 12:65–72. https:// doi. org/ 10. 1038/ 23. Silva AC, Lemos C, Goncalves FQ, Pliassova AV, Machado NJ, Silva HB, nrn29 67 Canas PM, Cunha RA, Lopes JP, Agostinho P (2018) Blockade of adenosine 5. Laurent C, Buée L, Blum D (2018) Tau and neuroinflammation: what A2A receptors recovers early deficits of memory and plasticity in the impact for Alzheimer’s disease and tauopathies? Biomed J 41:21–33. https:// doi. org/ 10. 1016/j. bj. 2018. 01. 003 Chang  et al. acta neuropathol commun (2021) 9:112 Page 17 of 18 triple transgenic mouse model of Alzheimer’s disease. Neurobiol Dis 40. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, 117:72–81. https:// doi. org/ 10. 1016/j. nbd. 2018. 05. 024 Salzberg SL, Rinn JL, Pachter L (2012) Differential gene and transcript 24. Viana da Silva S, Haberl MG, Zhang P, Bethge P, Lemos C, Goncalves N, expression analysis of RNA‑seq experiments with TopHat and Cufflinks. Gorlewicz A, Malezieux M, Goncalves FQ, Grosjean N et al (2016) Early Nat Protoc 7:562–578. https:// doi. org/ 10. 1038/ nprot. 2012. 016 synaptic deficits in the APP/PS1 mouse model of Alzheimer’s disease 41. Nolte H, MacVicar TD, Tellkamp F, Kruger M (2018) Instant clue: a software involve neuronal adenosine A2A receptors. Nat Commun 7:11915. suite for interactive data visualization and analysis. Sci Rep 8:12648. https:// doi. org/ 10. 1038/ ncomm s11915https:// doi. org/ 10. 1038/ s41598‑ 018‑ 31154‑6 25. Fang EF, Hou Y, Palikaras K, Adriaanse BA, Kerr JS, Yang B, Lautrup S, 42. da Huang W, Sherman BT, Lempicki RA (2009) Bioinformatics enrichment Hasan‑ Olive MM, Caponio D, Dan X et al (2019) Mitophagy inhibits tools: paths toward the comprehensive functional analysis of large gene amyloid‑β and tau pathology and reverses cognitive deficits in models lists. Nucleic Acids Res 37:1–13. https:// doi. org/ 10. 1093/ nar/ gkn923 of Alzheimer’s disease. Nat Neurosci 22:401–412. https:// doi. org/ 10. 1038/ 43. da Huang W, Sherman BT, Lempicki RA (2009) Systematic and integrative s41593‑ 018‑ 0332‑9 analysis of large gene lists using DAVID bioinformatics resources. Nat 26. Lauretti E, Li JG, Di Meco A, Praticò D (2017) Glucose deficit triggers tau Protoc 4:44–57. https:// doi. org/ 10. 1038/ nprot. 2008. 211 pathology and synaptic dysfunction in a tauopathy mouse model. Transl 44. Jicha GA, Bowser R, Kazam IG, Davies P (1997) Alz‑50 and MC‑1, a new Psychiatry 7:e1020. https:// doi. org/ 10. 1038/ tp. 2016. 296 monoclonal antibody raised to paired helical filaments, recognize confor ‑ 27. Ma T, Chen Y, Vingtdeux V, Zhao H, Viollet B, Marambaud P, Klann E mational epitopes on recombinant tau. J Neurosci Res 48:128–132 (2014) Inhibition of AMP‑activated protein kinase signaling alleviates 45. Mellone M, Kestoras D, Andrews MR, Dassie E, Crowther RA, Stokin GB, impairments in hippocampal synaptic plasticity induced by amyloid β. Tinsley J, Horne G, Goedert M, Tolkovsky AM et al (2013) Tau pathology J Neurosci 34:12230–12238. https:// doi. org/ 10. 1523/ jneur osci. 1694‑ 14. is present in vivo and develops in vitro in sensory neurons from human 2014 P301S tau transgenic mice: a system for screening drugs against tauopa‑ 28. Vingtdeux V, Davies P, Dickson DW, Marambaud P (2011) AMPK is abnor‑ thies. J Neurosci 33:18175–18189. https:// doi. org/ 10. 1523/ JNEUR OSCI. mally activated in tangle‑ and pre ‑tangle ‑bearing neurons in Alzheimer’s 4933‑ 12. 2013 disease and other tauopathies. Acta Neuropathol 121:337–349. https:// 46. Martin L, Latypova X, Wilson CM, Magnaudeix A, Perrin ML, Terro F (2013) doi. org/ 10. 1007/ s00401‑ 010‑ 0759‑x Tau protein phosphatases in Alzheimer’s disease: the leading role of 29. Herzig S, Shaw RJ (2018) AMPK: guardian of metabolism and mitochon‑ PP2A. Ageing Res Rev 12:39–49. https:// doi. org/ 10. 1016/j. arr. 2012. 06. 008 drial homeostasis. Nat Rev Mol Cell Biol 19:121–135. https:// doi. org/ 10. 47. Sayama S, Song A, Brown BC, Couturier J, Cai X, Xu P, Chen C, Zheng Y, 1038/ nrm. 2017. 95 Iriyama T, Sibai B et al (2020) Maternal erythrocyte ENT1‑mediated AMPK 30. Turnley AM, Stapleton D, Mann RJ, Witters LA, Kemp BE, Bartlett PF (1999) activation counteracts placental hypoxia and supports fetal growth. JCI Cellular distribution and developmental expression of AMP‑activated Insight. https:// doi. org/ 10. 1172/ jci. insig ht. 130205 protein kinase isoforms in mouse central nervous system. J Neurochem 48. Domise M, Didier S, Marinangeli C, Zhao H, Chandakkar P, Buee L, Viollet 72:1707–1716. https:// doi. org/ 10. 1046/j. 1471‑ 4159. 1999. 721707.x B, Davies P, Marambaud P, Vingtdeux V (2016) AMP‑activated protein 31. Medina‑Pulido L, Molina‑Arcas M, Justicia C, Soriano E, Burgaya F, Planas kinase modulates tau phosphorylation and tau pathology in vivo. Sci Rep AM, Pastor‑Anglada M (2013) Hypoxia and P1 receptor activation regulate 6:26758. https:// doi. org/ 10. 1038/ srep2 6758 the high‑affinity concentrative adenosine transporter CNT2 in differenti‑ 49. Horman S, Browne G, Krause U, Patel J, Vertommen D, Bertrand L, ated neuronal PC12 cells. Biochem J 454:437–445. https:// doi. org/ 10. Lavoinne A, Hue L, Proud C, Rider M (2002) Activation of AMP‑activated 1042/ BJ201 30231 protein kinase leads to the phosphorylation of elongation factor 2 and an 32. Pastor‑Anglada M, Errasti‑Murugarren E, Aymerich I, Casado FJ (2007) inhibition of protein synthesis. Curr Biol 12:1419–1423. https:// doi. org/ 10. Concentrative nucleoside transporters (CNTs) in epithelia: from absorp‑1016/ s0960‑ 9822(02) 01077‑1 tion to cell signaling. J Physiol Biochem 63:97–110. https:// doi. org/ 10. 50. Li S, Xiong GJ, Huang N, Sheng ZH (2020) The cross‑talk of energy sensing 1007/ BF031 74089 and mitochondrial anchoring sustains synaptic efficacy by maintaining 33. Schindowski K, Bretteville A, Leroy K, Begard S, Brion JP, Hamdane M, Buee presynaptic metabolism. Nat Metab 2:1077–1095. https:// doi. org/ 10. L (2006) Alzheimer’s disease‑like tau neuropathology leads to memory 1038/ s42255‑ 020‑ 00289‑0 deficits and loss of functional synapses in a novel mutated tau transgenic 51. Woods A, Dickerson K, Heath R, Hong SP, Momcilovic M, Johnstone SR, 2+ mouse without any motor deficits. Am J Pathol 169:599–616. https:// doi. Carlson M, Carling D (2005) Ca /calmodulin‑ dependent protein kinase org/ 10. 2353/ ajpath. 2006. 060002 kinase‑beta acts upstream of AMP ‑activated protein kinase in mamma‑ 34. Carvalho K, Faivre E, Pietrowski MJ, Marques X, Gomez‑Murcia V, Deleau lian cells. Cell Metab 2:21–33. https:// doi. org/ 10. 1016/j. cmet. 2005. 06. 005 A, Huin V, Hansen JN, Kozlov S, Danis C et al (2019) Exacerbation of C1q 52. DuBoff B, Gotz J, Feany MB (2012) Tau promotes neurodegeneration via dysregulation, synaptic loss and memory deficits in tau pathology linked DRP1 mislocalization in vivo. Neuron 75:618–632. https:// doi. org/ 10. to neuronal adenosine A2A receptor. Brain 142:3636–3654. https:// doi. 1016/j. neuron. 2012. 06. 026 org/ 10. 1093/ brain/ awz288 53. Cha MY, Cho HJ, Kim C, Jung YO, Kang MJ, Murray ME, Hong HS, Choi 35. Huin V, Deramecourt V, Caparros‑Lefebvre D, Maurage CA, Duyckaerts C, YJ, Choi H, Kim DK et al (2015) Mitochondrial ATP synthase activity is Kovari E, Pasquier F, Buée‑Scherrer V, Labreuche J, Behal H et al (2016) The impaired by suppressed O‑ GlcNAcylation in Alzheimer’s disease. Hum MAPT gene is differentially methylated in the progressive supranuclear Mol Genet 24:6492–6504. https:// doi. org/ 10. 1093/ hmg/ ddv358 palsy brain. Mov Disord 31:1883–1890. https:// doi. org/ 10. 1002/ mds. 54. Zhang Y, Chen K, Sloan SA, Bennett ML, Scholze AR, O’Keeffe S, Phatnani 26820 HP, Guarnieri P, Caneda C, Ruderisch N et al (2014) An RNA‑sequencing 36. Chang CP, Lee CT, Hou WH, Lin MS, Lai HL, Chien CL, Chang C, Cheng transcriptome and splicing database of glia, neurons, and vascular cells of PL, Lien CC, Chern Y (2016) Type VI adenylyl cyclase negatively regulates the cerebral cortex. J Neurosci 34:11929–11947. https:// doi. org/ 10. 1523/ GluN2B‑mediated LTD and spatial reversal learning. Sci Rep 6:22529. JNEUR OSCI. 1860‑ 14. 2014 https:// doi. org/ 10. 1038/ srep2 2529 55. Keren‑Shaul H, Spinrad A, Weiner A, Matcovitch‑Natan O, Dvir ‑Szternfeld 37. Van der Jeugd A, Ahmed T, Burnouf S, Belarbi K, Hamdame M, Grosjean R, Ulland TK, David E, Baruch K, Lara‑Astaiso D, Toth B et al (2017) A ME, Humez S, Balschun D, Blum D, Buee L et al (2011) Hippocampal unique microglia type associated with restricting development of Alzhei‑ tauopathy in tau transgenic mice coincides with impaired hippocampus‑ mer’s disease. Cell 169(1276–1290):e1217. https:// doi. org/ 10. 1016/j. cell. dependent learning and memory, and attenuated late‑phase long‑term 2017. 05. 018 depression of synaptic transmission. Neurobiol Learn Mem 95:296–304. 56. Liddelow SA, Guttenplan KA, Clarke LE, Bennett FC, Bohlen CJ, Schirmer https:// doi. org/ 10. 1016/j. nlm. 2010. 12. 005 L, Bennett ML, Munch AE, Chung WS, Peterson TC et al (2017) Neuro‑ 38. Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: a flexible trimmer for toxic reactive astrocytes are induced by activated microglia. Nature Illumina sequence data. Bioinformatics 30:2114–2120. https:// doi. org/ 10. 541:481–487. https:// doi. org/ 10. 1038/ natur e21029 1093/ bioin forma tics/ btu170 57. Gratuze M, Leyns CE, Sauerbeck AD, St‑Pierre MK, Xiong M, Kim N, Remo ‑ 39. Kim D, Langmead B, Salzberg SL (2015) HISAT: a fast spliced aligner with lina Serrano J, Tremblay ME, Kummer TT, Colonna M et al (2020) Impact of low memory requirements. Nat Methods 12:357–360. https:// doi. org/ 10. TREM2R47H variant on tau pathology‑induced gliosis and neurodegen‑ 1038/ nmeth. 3317 eration. J Clin Invest. https:// doi. org/ 10. 1172/ JCI13 8179 Chang et al. acta neuropathol commun (2021) 9:112 Page 18 of 18 58. Chatterjee S, Cassel R, Schneider‑Anthony A, Merienne K, Cosquer B, 67. Wang H, Li Y, Ryder JW, Hole JT, Ebert PJ, Airey DC, Qian HR, Logsdon B, Tzeplaeff L, Halder Sinha S, Kumar M, Chaturbedy P, Eswaramoorthy M Fisher A, Ahmed Z et al (2018) Genome‑ wide RNAseq study of the molec‑ et al (2018) Reinstating plasticity and memory in a tauopathy mouse ular mechanisms underlying microglia activation in response to patho‑ model with an acetyltransferase activator. EMBO Mol Med. https:// doi. logical tau perturbation in the rTg4510 tau transgenic animal model. Mol org/ 10. 15252/ emmm. 20170 8587 Neurodegener 13:65. https:// doi. org/ 10. 1186/ s13024‑ 018‑ 0296‑y 59. Clarke LE, Liddelow SA, Chakraborty C, Munch AE, Heiman M, Barres BA 68. Li Q, Cheng Z, Zhou L, Darmanis S, Neff NF, Okamoto J, Gulati G, Bennett (2018) Normal aging induces A1‑like astrocyte reactivity. Proc Natl Acad ML, Sun LO, Clarke LE et al (2019) Developmental heterogeneity of micro‑ Sci U S A 115:E1896–E1905. https:// doi. org/ 10. 1073/ pnas. 18001 65115 glia and brain myeloid cells revealed by deep single‑ cell RNA sequenc‑ 60. Bi F, Huang C, Tong J, Qiu G, Huang B, Wu Q, Li F, Xu Z, Bowser R, Xia XG ing. Neuron 101:207‑223.e210. https:// doi. org/ 10. 1016/j. neuron. 2018. 12. et al (2013) Reactive astrocytes secrete lcn2 to promote neuron death. 006 Proc Natl Acad Sci U S A 110:4069–4074. https:// doi. org/ 10. 1073/ pnas. 69. Dejanovic B, Huntley MA, De Mazière A, Meilandt WJ, Wu T, Srinivasan K, 12184 97110 Jiang Z, Gandham V, Friedman BA, Ngu H et al (2018) Changes in the syn‑ 61. Kerr JS, Adriaanse BA, Greig NH, Mattson MP, Cader MZ, Bohr VA, Fang EF aptic proteome in tauopathy and rescue of tau‑induced synapse loss by (2017) Mitophagy and Alzheimer’s disease: cellular and molecular mecha‑ C1q antibodies. Neuron 100:1322‑1336.e1327. https:// doi. org/ 10. 1016/j. nisms. Trends Neurosci 40:151–166. https:// doi. org/ 10. 1016/j. tins. 2017. 01. neuron. 2018. 10. 014 002 70. Gunner G, Cheadle L, Johnson KM, Ayata P, Badimon A, Mondo E, Nagy 62. Zheng J, Akbari M, Schirmer C, Reynaert ML, Loyens A, Lefebvre B, Buee L, MA, Liu L, Bemiller SM, Kim KW et al (2019) Sensory lesioning induces Croteau DL, Galas MC, Bohr VA (2020) Hippocampal tau oligomerization microglial synapse elimination via ADAM10 and fractalkine signaling. Nat early in tau pathology coincides with a transient alteration of mitochon‑ Neurosci 22:1075–1088. https:// doi. org/ 10. 1038/ s41593‑ 019‑ 0419‑y drial homeostasis and DNA repair in a mouse model of tauopathy. Acta 71. Habib N, McCabe C, Medina S, Varshavsky M, Kitsberg D, Dvir‑Szternfeld R, Neuropathol Commun 8:25. https:// doi. org/ 10. 1186/ s40478‑ 020‑ 00896‑8 Green G, Dionne D, Nguyen L, Marshall JL et al (2020) Disease‑associated 63. Domise M, Sauve F, Didier S, Caillerez R, Begard S, Carrier S, Colin M, astrocytes in Alzheimer’s disease and aging. Nat Neurosci 23:701–706. Marinangeli C, Buee L, Vingtdeux V (2019) Neuronal AMP‑activated https:// doi. org/ 10. 1038/ s41593‑ 020‑ 0624‑8 protein kinase hyper‑activation induces synaptic loss by an autophagy‑ 72. Lee MR, Ruby CL, Hinton DJ, Choi S, Adams CA, Young Kang N, Choi DS mediated process. Cell Death Dis 10:221. https:// doi. org/ 10. 1038/ (2013) Striatal adenosine signaling regulates EAAT2 and astrocytic AQP4 s41419‑ 019‑ 1464‑x expression and alcohol drinking in mice. Neuropsychopharmacology 64. Peng W, Wu Z, Song K, Zhang S, Li Y, Xu M (2020) Regulation of sleep 38:437–445. https:// doi. org/ 10. 1038/ npp. 2012. 198 homeostasis mediator adenosine by basal forebrain glutamatergic neu‑ rons. Science. https:// doi. org/ 10. 1126/ scien ce. abb05 56 Publisher’s Note 65. Gomez‑Murcia V, Sandau U, Ferry B, Parrot S, Laurent C, Basquin M, Buee Springer Nature remains neutral with regard to jurisdictional claims in pub‑ L, Boison D, Blum D (2020) Hyperexcitability and seizures in the THY‑ lished maps and institutional affiliations. Tau22 mouse model of tauopathy. Neurobiol Aging 94:265–270. https:// doi. org/ 10. 1016/j. neuro biola ging. 2020. 06. 004 66. Ho SY, Chen IC, Chang KC, Lin HR, Tsai CW, Lin CJ, Liou HH (2020) Equili‑ brative nucleoside transporters‑1 inhibitors act as anti‑ epileptic agents by inhibiting glutamatergic transmission. Front Neurosci 14:610898. https:// doi. org/ 10. 3389/ fnins. 2020. 610898 Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals

Loading next page...
 
/lp/springer-journals/equilibrative-nucleoside-transporter-1-inhibition-rescues-energy-ODIS0I0jXE
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2051-5960
DOI
10.1186/s40478-021-01213-7
Publisher site
See Article on Publisher Site

Abstract

Tau pathology is instrumental in the gradual loss of neuronal functions and cognitive decline in tauopathies, including Alzheimer’s disease (AD). Earlier reports showed that adenosine metabolism is abnormal in the brain of AD patients while consequences remained ill‑ defined. Herein, we aimed at investigating whether manipulation of adenosine tone would impact Tau pathology, associated molecular alterations and subsequent neurodegeneration. We demonstrated that treatment with an inhibitor (J4) of equilibrative nucleoside transporter 1 (ENT1) exerted beneficial effects in a mouse model of Tauopathy. Treatment with J4 not only reduced Tau hyperphosphorylation but also rescued memory deficits, mitochondrial dysfunction, synaptic loss, and abnormal expression of immune ‑related gene signatures. These beneficial effects were particularly ascribed to the ability of J4 to suppress the overactivation of AMPK (an energy reduction sensor), suggesting that normalization of energy dysfunction mitigates neuronal dysfunctions in Tauopathy. Collectively, these data highlight that targeting adenosine metabolism is a novel strategy for tauopathies. Keywords: Alzheimer’s disease, Tauopathy, Adenosine, AMPK, ENT1 Background 30 kinases and protein phosphatases [6, 7], thus making Alzheimer’s disease (AD) is the most prominent neuro- it very sensitive to the environment including metabolic degenerative disease in aging societies, but there is no changes [8, 9]. effective treatment [1]. The major pathogenic hallmarks Adenosine is an important homeostatic building block of Alzheimer’s disease include extracellular amyloid of many important metabolic pathways. It also serves as a plaques (amyloid-beta, Aβ) and intracellular accumula- neuromodulator that controls multiple functions (includ- tion of neurofibrillary tangles made of hyperphosphoryl - ing neuroinflammation, blood–brain barrier  permeabil - ated Tau. The latter is particularly known to be associated ity, neuronal transmission, and energy balance) through with neuritic dystrophy, synapse loss, and neuroinflam - receptor-dependent and/or independent mechanisms in mation, which lead to cognitive impairments [2–5]. the central nervous system [10, 11]. The main sources of Phosphorylation of Tau can be modulated by more than adenosine include ATP catabolism and the transmethyla- tion pathway. In addition, the extracellular and intracellu- lar adenosine levels are modulated through equilibrative *Correspondence: david.blum@inserm.fr; bmychern@ibms.sinica.edu.tw nucleoside transporters (ENTs) and concentrative nucle- Institute of Biomedical Sciences, Academia Sinica, Nankang, Taipei 115, Taiwan oside transporters [12, 13]. ENTs are bidirectional Univ. Lille, Inserm, CHU Lille, U1172 ‑ LilNCog ‑ Lille Neuroscience & transporter that transports adenosine in a concentra- Cognition, 59000 Lille, France tion-dependent, Na -independent manner. These ENT Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Chang et al. acta neuropathol commun (2021) 9:112 Page 2 of 18 members contain 11 transmembrane domains, and adenosine homeostasis modulation by J4 in a Tauopa- can be found in most cell types (including neurons and thy context, using the THY-Tau22 (Tau22) model, which astrocytes) in the brain. Among the four ENTs, ENT1 progressively develops hippocampal Tau pathology and has attracted much attention in recent years because it memory deficits [33]. We demonstrated that chronic has the highest affinity for adenosine [14, 15]. Since dis- treatment with J4 mitigates the development of hip- ruption of adenosine homeostasis in the brain has been pocampal Tau pathologies (including synapse loss, mito- implicated in several neurological disorders (including chondrial dysfunction, and neuroinflammation). These sleep disorders, impaired cognition impairment and neu- beneficial effects of J4 are particularly ascribed to its abil - rodegenerative diseases [16, 17]), modulation of adeno- ity to suppress AMPK overactivation in the hippocampi sine levels by controlling the components of adenosine of Tau22 mice, suggesting that it normalizes energy dys- metabolism therefore is a potential therapeutic approach. function caused by pathogenic Tau. Impaired adenosine homeostasis has been also sug- gested in the brain of AD patients [18]. In this study, Methods Alonso-Andres et  al. had reported that the adenosine Animals and drug administration levels in the frontal cortices of AD patients were lower THY-Tau22 (Tau22) mice ((B6.Cg-Tg(y1-Th than those of age-matched controls throughout disease MAPT)22Schd) expressing mutant human 4R Tau progression. Conversely, the adenosine levels of parietal (G272V and P301S) driven by a neuron-specific pro - cortices gradually become higher than normal subjects at moter (y1.2) Th were maintained on the C57BL/6J back - the later stage of AD [18]. No data of the hippocampal ground [33]. All animal studies were conducted following adenosine level was reported in this study. These findings the protocols approved by the Institutional Animal Care overall suggest that adenosine homeostasis may be dys- and Utilization Committee (IACUC, Academia Sinica, regulated in AD in a brain area-specific manner. Accord - Taiwan). All mice were housed in ventilated cages (IVC) ingly, dysregulation of the two main adenosine receptors, with freely accessible water and chow diet (LabDiet , A and A have been reported in the brain of patients San Antonio, TX, USA), and kept under controlled light- 1 2A [19, 20] as well as in AD mouse models [21–24]. ing condition with 12:12  h light/dark cycle at the Insti- The link between adenosine homeostasis and Tau tute of Biomedical Sciences Animal Care Facility (Taipei, pathology remains ill-defined. Interestingly, previous Taiwan). To investigate the effect of J4, male Tau22 mice studies have associated AMPK activation with abnor- and their littermate controls were randomly allocated to mal Tau phosphorylation in the brains of patients and experimental groups that treated with the J4 (0.02 mg/ml mice with Alzheimer’s disease [25–28]. AMP-activated in 1% HPβCD; designated TauJ and WTJ mice) or vehi- protein kinase (AMPK) is a homeostatic energy sensor cle (1% HPβCD; designated TauC and WTC mice) con- that controls the balance between anabolic and catabolic tinuously in drinking water for 7  months from the age processes in cells [29]. In the presence of stress (e.g., of 3 to 10  months. Males were chosen for the following an elevated AMP/ATP ratio, high reactive oxygen spe- experiments. The behavioral tests, electrophysiological cies levels, or mitochondrial dysfunction), AMPK can study, and RNA-seq analysis were carried out at the age be activated by being phosphorylated at Thr on the of 10  months. The immunohistochemical staining and α subunit of AMPK [30]. Under adverse conditions, for quantitative PCR were performed at age of 11  months. instance when the extracellular adenosine concentration The phospho-proteomic analysis was assessed at age of is elevated, transport of adenosine into cells enhances 12 months. Mice had continuously received the indicated the cellular level of AMP, alters the AMP/ATP ratio, and treatment during experiments. subsequently activates AMPK [31, 32]. Together, these observations raised the possibility of a link between Human cases adenosine homeostasis, AMPK and Tau. For immunofluorescence analysis, a total of 18 post-mor - We have addressed this link using the in-house syn- tem Human posterior hippocampal specimens: six nor- thetic compound J4. J4 is an orally active, BBB-perme- mal subjects, six Alzheimer’s disease, and six FTD-Tau able inhibitor of ENT1 with a Ki value of 0.05  μM [22]. (CBD, PSP, and Pick’s), were obtained from the UC Davis Intrahippocampal acute infusion of J4 elevated the extra- Alzheimer’s Disease Center (USA). cellular adenosine level determined by microdialysis in For mRNA analysis, a total of 55 post-mortem Human THY-Tau22 and WT mice (Additional file  1: Fig. S1). Its brain samples (Brodmann area 10 prefrontal cortex or oral bioavailability is 48% in mice with brain-to-blood temporal cortex): 13 normal subjects, 19 Alzheimer’s ratio of approximately 16.4% (Additional file  1: Table S1), Disease, and 23 FTD-Tau (CBD, P301L, PSP, and Pick’s), indicating that J4 can enter the brain via oral administra- were obtained from the brain banks of Lille, Paris, and tion. In the present study, we investigated the impact of Geneva. Participants and methods have been described Chang  et al. acta neuropathol commun (2021) 9:112 Page 3 of 18 previously [34, 35]. Fresh frozen grey matter tissue (about cerebrospinal fluid (ACSF; 119  mM NaCl, 2.5  mM KCl, 100  mg) retrieved at autopsy and stored at − 80  °C was 2.5 mM CaCl , 1.3  mM M gSO , 1  mM N aH PO , 2 4 2 4 used for mRNA analysis. All the brain samples used for 26.2 mM NaHCO , and 11 mM glucose) oxygenated with RT-qPCR analyses had an RNA integrity number ≥ 5. 95% O and 5% CO . Transverse hippocampal slices of 2 2 Detailed information on the normal subjects, Alzhei- 450 μm thickness were prepared with a DSK Microslicer mer’s Disease patients, and FTD-Tau patients from which (DTK-1000, Osaka, Japan) filled with oxygenated ice- the specimens used in the present study were obtained is cold ACSF. For recovery, the slices were then incubated listed in Additional file 1: Table S2. in an interface-type holding chamber filled with oxygen - ated ACSF at RT for at least 3  h. Before recording, the Morris water maze recording electrodes were prepared with a glass micro- Spatial memory and cognitive flexibility of mice aged pipette puller (PC-10, Narishige, Tokyo, Japan), and the 10–11  months were evaluated using the Morris water slices were transferred to an immersion-type recording maze test as described with slight modifications [36]. A chamber equipped with a perfusion system (flow rate: circular swimming pool (154  cm in diameter, 51  cm in 2–3  ml/min) and temperature controller (kept at 32  °C). height) was filled with milky water (30 cm in depth, kept To record field excitatory postsynaptic potentials (fEP - at 20  °C), and divided into four quadrants (T: target; L: SPs) recording, the bipolar stainless-steel stimulating target left; R: target right; O: opposite) with distinct electrodes (Frederick Haer Company, Bowdoinham, ME; visual cues on the tank wall of each quadrant. The hid - 10 ΩM impedance) and a glass pipette filled with 3  M den platform (13  cm in diameter, 0.5  cm below the sur- NaCl were placed in the stratum radiatum of the hip- face of the milky water) was placed in the center of the pocampal CA1 region. Basal synaptic transmission at target quadrant (T). Each mouse underwent four daily the Schaffer collateral-CA1 synapses was first evaluated training trial (120 s/trial, 30 min interval), in which they by measuring input–output curves using 12 stimuli (con- were released from randomly selected nontarget quad- stant current pulses from 10 to 120 μA in increments of rants (NTs). For the spatial memory test in the acquisi- 10 μA, duration of 40  μs). To measure the paired-pulse tion-learning phase, learning trials were performed with facilitation (PPF) response, two pulses were applied in a hidden platform for five consecutive days (Day 1–Day rapid succession (interpulse intervals of 50, 100, 150, 200, 5). For the spatial reversal memory test in the reversal- 300, 400 and 500 ms). Baseline responses were recorded learning phase, learning trials were performed with a by applying single stimuli (40 μs pulse-width) at 30  s hidden platform relocated to the opposite quadrant for intervals, and 4 responses were averaged to obtain a data an additional four consecutive days (Day 9–Day 12). To point. Long-term depression (LTD) was induced using 3 evaluate reference memory, the probe trial and rever- trains of low-frequency stimulation (LFS, 1200 pulses at sal probe trial were performed on Day 8 (72  h after the 2 Hz) with a 10-min interval/train as described elsewhere acquisition-learning phase) and Day 15 (72  h after the [37]. The initial fEPSP slope was calculated by using Sig - reversal-learning phase), respectively. For the probe test, nal software (V4.08, Cambridge Electronic Design, Cam- the hidden platform was first removed. The mouse was bridge, UK). then released in the opposite quadrant (O), and their swimming path was recorded for 120  s. The swimming Phospho‑proteomic analysis path and other parameters (e.g. escape latency and swim- Hippocampal lysates (200  μg) collected from three ani- ming speed) of each mouse in different quadrants were mals from each condition were subjected to phospho- monitored and analyzed using the TrackMot video track- proteomic analysis at the Proteomics Core Facility (PCF) ing system (Diagnostic & Research Instruments Co., Ltd., (Institute of Biomedical Sciences, Academia Sinica, Tai- Taoyuan, Taiwan). Mice that exhibited nonsearching pei, Taiwan). For in-solution digestion, 8  M urea was behaviors (floating, a swimming speed below 10  cm/s added to the lysates to prepare a mixture of 10  mg pro- and circling) were excluded from the analysis. Statistical tein/ml in 6  M urea. Protein reduction was performed differences were analyzed by two-way ANOVA. by the addition of dithiothreitol (DTT, final concentra - tion 5  mM) and incubation at 56  °C for 25  min. Protein Electrophysiological study alkylation was performed by the addition of iodoaceta- Mice aged 10–11  months were used for electrophysi- mide (IAA, final concentration 15  mM) and incubation ology approaches. All electrophysiology studies were at RT for 45  min to block the reversion of sulfhydryl (– performed at the electrophysiology core facility (Neu- SH) groups to disulfide bonds. Trypsin/Lys-C Mix (Pro - roscience Program of Academia Sinica, Taipei, Tai- mega, WI, USA) was then added to protein (protein: wan). After rapid decapitation, the hippocampus was protease = 25:1 w/w) and the mixture was incubated for quickly dissected out and immersed in ice-cold artificial 4 h at 37 °C. The urea concentration was adjusted to 1 M Chang et al. acta neuropathol commun (2021) 9:112 Page 4 of 18 or less by diluting the reaction mixture with triethylam- normalization. References of the probes used in this monium bicarbonate (TEAB, 50  mM) followed by an study are given in Additional file 1: Table S4. incubation at 37  °C for 17  h. The digested samples were dried with a SpeedVac and desalted with C18 Oasis RNA sequencing (RNA‑seq) PRiME HLB cartridges (Waters, MA, USA). For iTRAQ Total RNA samples (3 μg per sample) extracted from the labeling, the digested peptides were labeled with four hippocampus with RIN values greater than 8 were sub- isobaric iTRAQ Reagents (114, 115, 116, and 117) using jected to RNA-seq analysis. The RNA library construc - iTRAQ Reagents-4plex Applications Kit (AB Sciex, tion and sequencing were carried out by Welgene Biotech MA, USA) following manufacturer’s instructions. For (Taipei, Taiwan). Briefly, the SureSelect Strand-Specific phosphopeptide enrichment, iTRAQ labeled peptides RNA Library Preparation Kit (Agilent Technology, CA, were mixed with loading buffer (80% acetonitrile, ACN, USA) was used for library construction on the Illumina 5% trifluoroacetic acid, TFA, and 1  M glycolic acid) and platform. After AMPure XP Bead-based (Beckman adjusted to pH 2. The sample solution was then mixed Coulter Genomics, MA, USA) size selection of the RNA with TiO beads (GL Sciences, Japan) and incubated with 2 library, the sequences were determined using the Illumi- vortexing at RT for 15 min. The beads were collected and na’s sequencing-by-synthesis (SBS) technology to obtain washed twice with 100 μl washing buffer (80% ACN and 150-bp paired-end reads. Sequencing data (FASTQ files) 5% TFA). The phosphopeptides were sequentially eluted were generated by Welgene’s pipeline (Base call conver- with 50 μl 0.5% N H OH, 50 μl 5% N H OH, and 50 μl 80% 4 4 sion, adaptor clipping, and sequence quality trimming) ACN with 0.1% formic acid, and dried with a SpeedVac. based on Illumina’s base-calling program bcl2fastq v2.2.0 The phosphopeptides were then subjected to LC/MS/MS (Illumina, CA, USA) and Trimmomatic v0.36 [38]. The and analyzed by Proteome Discoverer ver.2.2 (Thermo RNA-seq reads were then aligned to the mouse reference Fisher Scientific, Waltham, MA, USA). genome (mm10) from the Ensembl database (Ensembl release 93) using HISAT2 [39]. The sample-to-sample distances were visualized by principal components anal- RNA extraction, cDNA synthesis, and quantitative PCR ysis (PCA) (Additional file  1: Fig. S2). Expression levels For mouse brain tissue, RNA isolation and complemen- (fragments per kilobase per million, FPKM) were ana- tary DNA (cDNA) synthesis were performed according lyzed and estimated using cuffdiff (cufflinks v2.2.1) [40] to the manufacturer’s protocols. In brief, mouse hip- and Welgene in-house programs. To identify the differ - pocampal tissues were homogenized in GENEzol rea- entially expressed (DE) genes in different groups, cutoff gent (GZX100, Geneaid Biotech Ltd., New Taipei City, criteria (absolute lo g fold change ≥ 0.32, p < 0.05) were Taiwan) with sterilized tissue grinders (FocusBio, Tai- used. A volcano plot and heatmap of the DE genes were wan), and then standard procedures for RNA preparation drawn by using Instant Clue [41] and Morpheus software and cDNA synthesis were performed as described previ- (https:// softw are. broad insti tute. org/ morph eus/), respec- ously [22]. Quantitative PCR (qPCR) assays were carried tively. The Gene Ontology (GO) and the Kyoto Encyclo - out using the LightCycler 480 System (Roche Life Sci- pedia of Genes and Genomes (KEGG) pathways of DE ence, Indiana, USA) and analyzed by the comparative CT genes were analyzed by using the Database for Annota- (ΔΔCt) method with GAPDH as a reference gene. The tion, Visualization and Integrated Discovery (DAVID 6.7) sequences of the PCR primers are shown in Additional [42, 43]. The Ingenuity Pathway Analysis (IPA) software file 1: Table S3. (Qiagen, CA, USA) was then used for the identification For Human brain tissue, total mRNA was extracted and of canonical pathway(s) related to the DE genes. Z-score purified using the RNeasyLipid Tissue Mini Kit (Qiagen). was used to predict activation (Z-score ≥ 2.0) or inhibi- One microgram of total mRNA was reverse-transcribed tion (Z-score ≤ − 2.0) of the indicated pathway. using the HighCapacity cDNA reverse transcription kit (Applied Biosystems). Quantitative real-time polymer- Immunohistochemical staining ase chain reaction (qPCR) analysis was performed on Coronal brain sections  (20  μm) from the desired mice ™ ™ an Applied Biosystems StepOnePlus Real-Time PCR were prepared as previously described [22]. For immu- Systems using TaqMan Gene Expression Master Mix nofluorescence (IF) staining, brain slices were washed (Applied Biosystems ). The thermal cycler conditions with 0.1  M PBS buffer, permeabilized with 0.2% Triton were as follows: 95 °C for 10 min, then 40 cycles at 95 °C X-100 solution (in 0.1  M PBS buffer), and blocked with for 15  s and 60  °C for 1  min. Amplifications were car - 3% normal goat serum (NGS), 3% normal donkey serum ried out in duplicate and the relative expression of target (NDS), or 3% bovine serum albumin (BSA) in 0.1 M PBS genes was determined by the ΔΔCt method with β-Actin buffer for 2 h at RT. The brain sections were then washed (ACTB) was used as a reference housekeeping gene for Chang  et al. acta neuropathol commun (2021) 9:112 Page 5 of 18 with 0.1  M PBS buffer twice and incubated with the deaminase (ADA), CD39, CD73) involved in adeno- indicated primary antibodies (listed in Additional file  1: sine metabolism were found elevated in the hippocam- Table S5) in primary antibody solution (1% NGS or BSA, pus of Tau22 mice, supporting a change in adenosine 0.2% Triton X-100, and 0.1% sodium azide in 0.1 M PBS) homeostasis. A trend of increase in the transcript level for 48  h at 4  °C. After extensive washes, the brain sec- of adenosine kinase (ADK) was also found, but did not tions were incubated with the corresponding secondary reach statistical significance. Importantly, J4 treatment antibody (1:500) for 2  h at RT, and then the nuclei were normalized all these changes (Table  1), suggesting that stained with Hoechst 33258 (1:5000) for 10  min at RT. J4 was able to reinstate proper adenosine homeostasis Free-floating brain sections were mounted on the silane- in the hippocampus of Tau22 mice. coated slides (Muto Pure Chemicals Co., Tokyo, Japan) with mounting media (Vector Laboratories, CA, USA) and stored at 4 °C before imaging. An LSM 780 confocal Chronic J4 treatment prevents impairment of spatial microscope (Carl Zeiss, Germany) was used to capture learning and memory of Tau22 mice images. The images were analyzed with MetaMorph soft - Under the tested condition, the spatial memory and ware (Universal Imaging, PA, USA). cognitive flexibility of mice at the age of 10  months Formalin-fixed, paraffin-embedded human brain slices were evaluated using the Morris water maze (MWM) were deparaffinized and rehydrated. To expose the anti - task. J4 significantly prevented spatial learning defi - genic sites, the brain slices were then immersed in 1X cits in vehicle-treated Tau22 mice (TauC) during the citrate buffer (C9999, Sigma-Aldrich, St. Louis, MO, acquisition-learning phase (Day 1–Day 5) (Fig.  1a) USA) for 20  min at 97.5  °C and cooled to RT. The brain without affecting wild-type mice. Moreover, vehicle-, slices were subjected to immunofluorescence staining J4-treated WT, and J4-treated Tau22, but not vehicle- as described above. After secondary antibody (1:500) treated Tau22 mice showed a preference for searching incubation, the brain sections were treated with 0.1% the hidden platform in the target quadrant (T) over the (w/v) Sudan Black B (199664, Sigma) in 70% ethanol for nontarget quadrants (NTs) in the probe test (Fig.  1b), 15 min at RT to block autofluorescence signal. The brain indicating that J4 improved the impaired memory of sections were then stained with Hoechst 33258 (1:5000) Tau22 mice. for 10  min at RT. After extensive washes with PBS, the To assess cognitive flexibility, we also examined the brain sections were mounted with the mounting media spatial reversal learning of WT and Tau22 mice treated (Vector Laboratories, CA, USA) and stored at 4 °C before with or without J4 for an additional four consecutive imaging. days. In the reversal-learning phase (Day 9–Day 12), the hidden platform was relocated to the opposite quadrant and the escape latency was recorded. J4 improved the Statistics impaired spatial reversal learning of TauC mice (Fig. 1a). The experimental condition was blinded to investigators Animals were subjected to the probe test on Day 15. during behavioral and electrophysiological experiments. Except for vehicle-treated Tau22 mice, all mice spent The data are expressed as the mean ± S.E.M. All statisti- more time in the new target quadrant (T) than in the cal analyses were performed using GraphPad Prism Soft- nontarget quadrants (NTs) (Fig. 1c). No difference in the ware (La Jolla, CA, USA). Two-tailed unpaired Student’s swimming speed was observed among the groups tested t-test was used to compare the difference between the two groups. One-way or two-way ANOVA followed by Tukey’s multiple comparisons test was used for compari- son of multiple groups. Differences were considered sta - tistically significant when p < 0.05. Table 1 Enzymes involved in adenosine homeostasis Gene WTC WTJ TauC TauJ Results Chronic J4 treatment stabilizes the abnormally altered # ADA 1.06 ± 0.09 1.04 ± 0.06 3.47 ± 0.42* 1.35 ± 0.24 expressions of genes that control adenosine metabolism # ADK 1.02 ± 0.03 1.06 ± 0.02 1.17 ± 0.06 0.85 ± 0.09 WT and Tau22 male mice were treated with J4 # CD39 1.01 ± 0.03 1.23 ± 0.02* 1.45 ± 0.06* 0.94 ± 0.05 (0.02  mg/ml in 1% HPβCD) or vehicle (1% HPβCD) in # CD73 1.02 ± 0.04 1.23 ± 0.07* 1.33 ± 0.05* 1.13 ± 0.05 drinking water for 7 months from the age of 3 months. Mice were treated as indicated (control WT mice, WTC; J4-treated WT mice, The average daily intake level of J4 was 3.06 ± 0.28  mg/ WTJ; control Tau22 mice, TauC; and J4-treated Tau22 mice, TauJ; n = 6–9 in kg. The hippocampi of the treated animals were then each group) from the age of 3–11 months. The hippocampus was harvested carefully and subjected to RT-qPCR. GAPDH was used as a reference gene for collected to harvest total RNA for RT-qPCR analy- normalization. The data are expressed as the mean ± SEM. *p < 0.05 versus the sis. The levels of at least three genes (i.e., adenosine WT vehicle group; p < 0.05 versus the Tau22 vehicle group Chang et al. acta neuropathol commun (2021) 9:112 Page 6 of 18 (Fig.  1d). Together, J4 significantly improves the spatial Alzheimer’s Disease was further assessed by Western memory and spatial reversal memory of Tau22 mice. blot and immunofluorescence analysis using antibodies raised against hyperphosphorylated (pThr181, pSer199, Chronic J4 treatment normalizes the impairment pSer202/Thr205 = AT8, pSer262, pSer396, and pSer422) of hippocampal CA1 LTD in Tau22 mice and misfolded (pThr212/Ser214 = AT100; MC1) Tau. We next investigated whether J4 affects the synaptic plas - Chronic treatment with J4 significantly reduced Tau ticity in Tau22 mice. The basal transmission of the hip - phosphorylation at the sites tested except for pSer181 pocampal CA3–CA1 network was determined based (Figs.  2b, c, Additional file  1: S3A, and S3B). No effect on the input–output relationship. The vehicle-treated of J4 on the human Tau was observed, demonstrating Tau22 mice showed decreased basal synaptic transmis- that J4 did not modulate the Thy-1 promoter directly sion compared to that of the vehicle-treated WT mice. (Additional file  1: Fig. S3C–S3G). Importantly, the lev- Such abnormal synaptic transmission was alleviated by els of misfolded Tau, assessed by AT100 and MC1 [44, J4 in Tau22 mice (Fig.  1e). Presynaptic neurotransmitter 45], were also reduced by J4 (Fig.  2b, c). Collectively, J4 release in the hippocampus, as determined by the paired- reduces the levels of hyperphosphorylated and misfolded pulse facilitation (PPF) response assay, was comparable Tau in Tau22 hippocampi. between genotypes (Tau22 versus WT mice) and treat- ment groups (J4 versus vehicle) (Fig.  1f ), suggesting that Chronic J4 treatment reduces the AMPK activation presynaptic plasticity was unaffected in Tau22 mice. in the hippocampus of Tau22 mice Previous studies have demonstrated that Tau22 mice We next examined the phosphorylation level of 17 exhibit impaired long-term depression (LTD) but nor- kinases and signaling molecules by using the MAPK mal long-term potentiation (LTP) at the Schaffer collat - Phosphorylation Array (RayBiotech, GA, USA). Only eral-CA1 synapses in the hippocampus [33]. As shown minor or no change was found between Tau22 mice and in Fig.  1g, LTD was maintained in WTC mice but not in WT mice (Fig. S4). No marked alterations in the level or TauC mice. This impairment in LTD was prevented by activation of PP2A [46], the major tau phosphatase, were J4. The average LTD magnitude during the last 10  min detected either (Additional file 1: Fig. S5). of recording was quantified and shown in Fig.  1h. Col- Because adenosine homeostasis has been implicated in lectively, J4 normalizes the impaired basal synaptic trans- the regulation of AMPK activation [31, 32, 47] and since mission and LTD at Schaffer collateral synapses in Tau22 AMPK directly phosphorylates Tau [48], we evaluated hippocampi without affecting those in WT hippocampi. the activation/phosphorylation of AMPK (Thr , des- ignated pAMPK) and Tau phosphorylation in the hip- Chronic J4 treatment reduces Tau hyperphosphorylation pocampi of postmortem Alzheimer’s Disease, FTLD-Tau in the hippocampi of Tau22 mice patients, and Tau22 mice. Immunofluorescence staining To examine the effect of J4 on Tau phosphorylation [33], revealed that, in the posterior hippocampal sections from we performed a differential peptide  labeling  (iTRAQ) Alzheimer’s Disease and FTD-Tau patients, pAMPK was of hippocampal proteins and analyzed the results using detected in neurons that contained phosphorylated Tau LC–MS/MS-based proteomics. Phosphopeptides cover- (Figs.  3a and Additional file  1: S6). Similarly, elevated ing 30 phosphorylation sites of human Tau were iden- AMPK phosphorylation was observed in neurons con- tified (Fig.  2a). In total, J4 reduced phosphorylation at taining phosphorylated Tau in the CA1 region in Tau22 15 phosphorylation sites. The J4-mediated reduction mice, but rarely in those in WT mice (Fig. 3b, c). J4 treat- in Tau phosphorylation at sites commonly observed in ment decreased the levels of pAMPK and pTau in Tau22 (See figure on next page.) Fig. 1 Chronic treatment with J4 alleviates the impairment of spatial memory and hippocampal CA1 LTD in Tau22 mice. a–d Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green; n = 12–20 in each group) from the age of 3–10 months. The Morris water maze (MWM) with a hidden platform was used to assess spatial learning and memory. a The acquisition‑learning phase (Day 1–Day 5) and the reversal‑learning phase (Day 9–Day 12) of MWM. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group; two‑ way ANOVA. Probe tests for b the spatial memory and c spatial reversal memory were performed on Day 8 and Day 15, respectively. The average percentage of time spent in the T (target quadrant) and NTs (nontarget quadrants) were calculated. *p < 0.05 versus the NTs; two‑tailed Student’s t-test. d The swimming speed (cm/s) of the animals in the probe test. e–h Hippocampal slices were prepared from mice subjected to different treatment groups (n = 3–18 slices from 3 to 8 mice in each group) from the age of 3–10 months. e The input/output relationship curve (fEPSP responses, stimuli strengths increased from 10 to 120 μA), f averaged paired‑pulse ratios (interstimulus intervals, from 50 to 500 ms), and g, h Long‑term depression (LTD) induced by 3 trains of LFS (2 Hz, 1200 pulses) at the Schaffer collateral‑ CA1 synapses were recorded. g The average fEPSP slopes of mice subjected to different treatments were calculated. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group, two‑ way ANOVA. h Quantification results of the mean fEPSP slopes during the last 10 min of the steady‑state period. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group, one‑ way ANOVA. The data are expressed as means ± S.E.M Chang  et al. acta neuropathol commun (2021) 9:112 Page 7 of 18 Chang et al. acta neuropathol commun (2021) 9:112 Page 8 of 18 mice compared to WT mice (Fig. 3b–d). In line with this Tauopathy-associated genes. Gene ontology (GO; Fig. 4c, finding, analyses of hippocampal proteins using a com - FDR < 0.05, Benjamini p < 0.01) and Kyoto Encyclope- bination of iTRAQ and LC–MS/MS-based proteomics dia of Genes and Genomes (KEGG; Fig.  4d, FDR < 0.05, by Phosphopeptides revealed that the phosphorylation Benjamini p < 0.01) analyses were conducted, and the levels of two AMPK downstream targets (i.e., eukary- DE genes between the TauC group vs WTC group were otic elongation factor 2 (eEF2) and myosin VI (Myo6); found to be associated with the immune response and 2+ [49, 50]) and one upstream regulator (Ca /calmodulin- transcriptional machinery. Most of the pathways dereg- dependent protein kinase kinase-β, Camk2b; [51]) were ulated by Tau pathology (red bars in Fig.  4c, d) became elevated in Tau22 mice. J4 treatment reduced the phos- less significant after J4 treatment (green bars). When we phorylation of eEF2, Myo6, and Camk2b (Additional analyzed the upregulation- or downregulation-specific file  1: Table S6), supporting that J4 normalized the upreg- DE genes separately (Additional file  1: Fig. S7A–S7C and ulated AMPK signaling pathway in the hippocampus of S7D–S7F, respectively), multiple pathways (e.g., tran- Tau22 mice. scription-related machineries, angiogenesis, cell–cell Since AMPK is a critical energy sensor and Tau has interaction, cell adhesion) remained markedly normal- been implicated in mitochondrial dysfunction [52], we ized by J4 treatment. Part of the inflammation-related next assessed mitochondrial mass by immunohisto- pathways were also rescued by J4 treatment. chemical staining using an antibody against ATP5a, a Next, we specifically analyzed the effect of J4 on Tau22 component of complex V [53]. Consistent with abnormal mice vs. control Tau22 mice. A total of 1239 DE genes AMPK activation, TauC mice exhibited less ATP5a-posi- were identified (289 upregulated and 950 downregulated; tive mitochondrial mass in the hippocampus than WTC TauJ mice versus TauC mice, p < 0.05 and absolute log mice (Fig.  3e). J4 reversed mitochondrial loss in Tau22 fold-change ≥ 0.32; Fig.  4e). Importantly, J4 normalized mice (Fig.  3e, f ) as it normalized the AMPK overactiva- the expression of 436 of the 950 upregulated DE genes tion (Fig.  3b, c), suggesting that the blockade of ENT1 (45.8%; Fig.  4f) and 85 of the 491 downregulated DE normalized Tau-associated energy dysfunction. genes (17.3%, data not shown) between the TauC group and WTC group. Moreover, J4 normalized multiple dys- Transcriptomic signature associated with the beneficial regulated canonical pathways, which were also observed effect of J4 in the hippocampi of Tau22 mice in other Alzheimer’s disease mouse models (Tg4510 To gain mechanistic insight into J4’s action, we per- and APP/PS1), in Tau22 mice (Additional file  1: Fig. formed transcriptional profiling of the hippocampus S8). Overall, J4 had a broad impact on Alzheimer’s Dis- using RNA-seq analysis. A total of 1441 differentially ease-related signaling molecules and pathways in Tau22 expressed (DE) genes (950 upregulated and 491 down- hippocampi. regulated) between TauC and WTC mice were identified (Fig. 4a, TauC mice versus WTC mice, p < 0.05 and abso- Chronic J4 treatment mitigates the activation of microglia lute log fold-change ≥ 0.32). Between the TauJ and WTC in the hippocampi of Tau22 mice groups, a total of 1304 DE genes were also identified Based on the cell-type information listed in the Brain (Fig.  4b; TauJ mice versus WTC mice, p < 0.05, absolute RNAseq database (https:// www. brain rnaseq. org, [54]), log fold-change ≥ 0.32) but the proportion of upregu- we further classified the 436 DE genes with expres - lated (417) and downregulated (887) markedly differed. sion levels normalized by J4 (Fig.  4f) into five types Interestingly, only a fraction (216 upregulated; 214 down- (including neuron-enriched, glia-enriched, endothelial regulated; (Additional file  1: Fig. S7A and S7D) of the cell-enriched, and unclassified). As shown in Fig.  4g, genes affected in Tau mice as compared to WT, remained approximately 55% of the DE genes whose expression lev- altered in the Tau mice after J4 treatment. These data sug - els that were normalized by J4 were enriched in glial cells, gested that J4 significantly normalized the expressions of including astrocytes and microglia (Fig.  4g). In Tau22 (See figure on next page.) Fig. 2 Chronic J4 treatment decreases hyperphosphorylated and misfolded human Tau levels in the hippocampi of Tau22 mice. Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green) for 8–9 months from the age of 3 months. a Pooled total hippocampal lysates (200 μg) from 3 animals of the age of 12 months were harvested and subjected to phospho‑proteomic analysis. The heatmap shows the relative log2 expression ratio of phosphorylated human tau (MAPT, P10636) in the TauJ group vs. the TauC group. The relative expression level (log2 ratio) of human phosphorylated tau is shown on a scale from red (upregulated) to blue (downregulated). The identified phosphorylation sites on peptides are shown in bold and underlined. b, c Hippocampal sections (20 μm) were prepared from mice with different treatment groups (n = 3–5 in each group) from the age of 3–11 months and subjected to IHC staining. b The levels of hyperphosphorylated tau and misfolded tau in the hippocampus were evaluated by staining with the indicated antibodies (AT8 for Ser202/Thr205, green; AT100 for Thr212/Ser214, green; MC1 for conformational changed tau, green), and the quantification results are shown in (c). Scale bar, 50 μm. The data are expressed as the mean ± S.E.M. p < 0.05, versus the TauC group, two‑tailed Student’s t-test Chang  et al. acta neuropathol commun (2021) 9:112 Page 9 of 18 C Chang et al. acta neuropathol commun (2021) 9:112 Page 10 of 18 Fig. 3 Chronic J4 treatment decreases AMPK activation and rescues mitochondrial abnormalities in the hippocampi of Tau22 mice. a Posterior hippocampal sections (6 μm) from normal subjects and Alzheimer’s Disease and FTD‑ Tau (CBD, PSP, and Pick’s disease) patients were subjected to Thr172 IHC staining. The levels of phospho‑AMPK and hyperphosphorylated tau were evaluated by staining with the indicated antibodies (pAMPK , Ser202/Thr205 green; AT8 for pTau , red). b–f Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green; n = 5–7 in each group) from Thr172 the age of 3–11 months. Hippocampal sections (20 μm) were prepared and subjected to IHC staining using the indicated antibodies (pA MPK , Ser202/Thr205 green; AT8 for pTau , red), and the staining was quantified (c, d). Scale bar, 20 μm. e, f The level of the mitochondrial marker ATP5a was evaluated by staining with an anti‑ATP5a antibody (e, green) and quantified (f; n = 3 in each group). Scale bar, 5 μm. The data are expressed as the mean ± S.E.M. *p < 0.05 versus the WTC group; p < 0.05, versus the TauC group, one‑ way ANOVA hippocampi, enhanced gene signatures for the disease- mice (Fig.  5a, b). We also measured the transcript lev- associated microglia were detected (DAM, [55]; Fig.  4h els of several factors secreted by activated microglia and and Additional file  1: Table S7). Consistently, the immu- known to favor neurotoxic activation of astrocytes (A1 noreactive intensities of both Iba1 (a marker of micro- phenotype; [56]) using RT-qPCR. As shown in Fig.  5c, glia) and CD68 (a marker of activated microglia) were the levels of TNF-α and C1q (C1qa, C1qb, and C1qc), but significantly elevated in the hippocampi of TauC mice not IL-1α, were upregulated in the Tau22 hippocampi (11 months old) compared with the hippocampi of WTC compared with WT hippocampi. Immunofluorescence Chang  et al. acta neuropathol commun (2021) 9:112 Page 11 of 18 staining and RT-qPCR showed that J4 prevented the secondary to microglial activation in Tau22 hippocampi upregulation of TNF-α, C1q, and CD68 in Tau22 mice since activated microglia (CD68-positive) were detected (Fig.  5a–e, Additional file  1: S9A and S9B), suggesting in the hippocampi of young Tau22 mice (4  months that J4 mitigated microglial inflammation in Tau22 mice. old), when no reactive astrocytes (Lcn2-positive) were observed (Additional file 1: Fig. S11). Chronic J4 treatment mitigates synaptic loss in Tau22 mice According to the reduction of the microglial phenotype C1q is an important mediator of Tau-induced synaptic as well as TNFα and C1q expressions (Fig. 5c), chronic J4 loss [34, 57]. We thus examined the number of synapses treatment normalized not only GFAP and Lcn2 levels but by immunofluorescence staining. In Tau22 hippocampi, also the pathological upregulation of A1-specific genes the levels of a postsynaptic marker (PSD95) and a pre- expression (Figs.  4i, 5h, i; Additional file  1: Table  S8). synaptic marker (synaptophysin) were lower than those Collectively, our data suggest that early Tau-induced in WT hippocampi. Consistent with the rescuing effect microglial activation is likely to promote the activation of of J4 on C1q, J4 restored the levels of PSD95 and synap- neurotoxic astrocytes and can be blocked by J4. tophysin in Tau22 mice (Additional file  1: Fig. S9C and S9D). We also determined the number of synapses based Discussion on the colocalization of PSD95 and synaptophysin. In The present study showed that chronic treatment with J4, line with the reduction in CD68 and C1q levels (Fig. 5a– an ENT1 blocker, mitigates Tau pathology by alleviating c), J4 rescued the synaptic loss in Tau22 hippocampi [58] not only mitochondrial dysfunction and AMPK overac- (Fig. 5f, g). tivation but also the neuroinflammatory status of micro - glia and astroglia, ultimately attenuating the impairment Chronic J4 treatment suppresses the cytotoxic astrocytes of compromised synapses as well as spatial learning and induction in the hippocampi of Tau22 mice memory. Our study particularly supports a functional TNF-α and C1q are potent astrocytic activators for the link between adenosine homeostasis, AMPK regulation neurotoxic A1 phenotype [56]. Our RNA-seq analysis of and Tau pathology development. Tau22 hippocampi revealed the upregulation of gene sig- Mitochondrial dysfunction is a major pathogenic fea- natures of pan-reactive and cytotoxic A1 astrocytes ([59], ture of Alzheimer’s Disease [61] and is known to facili- Additional file  1: Table  S8 and Fig.  4i). Immunofluores - tate the hyperphosphorylation of Tau, which in turn cence staining further showed that Tau22 mice (TauC) alters the morphology and functions of mitochondria exhibited higher levels of GFAP (an astrocyte marker) [62]. Therefore, it is not surprising that AMPK, a key and Lcn2 (a pan reactive astrocyte marker, [60]) in their energy sensor and an upstream kinase of Tau, is over- hippocampus than WT mice (WTC; Fig. 5h, i), confirm - activated in the hippocampi of patients with Alzhei- ing that astrocytes in Tau22 hippocampi were abnormally mer’s Disease or tauopathies [28]. One major function activated. Interestingly, FTD-Tau patients and late stage of AMPK is the maintenance of cellular energy homeo- Alzheimer’s Disease patients (Additional file  1: Fig. S10), stasis through modulation of the balance between ana- upregulation of several A1-specific genes (e.g., GBP2, bolic and catabolic processes [29]. Because hippocampal SERPING1, FKBP5) was observed. Notably, the five A1 neurons of WT mice are homeostatic in nature, no sig- astrocyte genes tested were all significantly upregulated nificant AMPK activation was observed in the hippocam - in the frontal cortex of FTD-Tau-Pick’s disease patients pus of WT mice (Fig.  3b, c). Treatment with J4 showed compared to control subjects, suggesting a link between no impact on AMPK activation in such a homeostatic Tau and astrocyte reactivity (Additional file  1: Fig. condition, suggesting that ENT1 does not play a sig- S10). The induction of reactive A1 astrocytes appears nificant role in the regulation of AMPK in physiological (See figure on next page.) Fig. 4 Chronic J4 treatment ameliorates the expression of tauopathy associated genes and neurotoxic reactive astrocyte genes in the hippocampi of Tau22 mice. Mice were treated as indicated ( WTC, TauC, and TauJ; n = 3 in each group) from the age of 3–10 months. The hippocampus was carefully removed for RNA‑seq analysis. Volcano plot of DE genes in the a TauC group vs. WTC group and b TauJ group vs. WTC group. c GO enrichment analysis of DE genes between the TauC and WTC groups (red bar) and TauJ and WTC groups (green bar). d KEGG pathway analysis of DE genes between the TauC and WTC groups (red bar) and TauJ and WTC groups (green bar). e Volcano plot of the DE genes in the TauJ group vs. TauC group. In the volcano plot, the significantly upregulated and downregulated DE genes (absolute log ratio ≥ 0.32; p < 0.05) are shown in red and blue, respectively. f Venn diagram showing the 436 overlapping DE genes between the upregulated DE genes in the TauC group vs. WTC group (pink) and the downregulated DE genes in the TauJ group vs. TauC group (green). g Pie chart of cell‑type ‑ enrichment of 436 tauopathy‑associated DE genes regulated by J4. In the volcano plot, the significantly upregulated and downregulated DE genes (absolute log ratio ≥ 0.32; p < 0.05) are shown in red and blue, respectively. In the Venn diagram and pie chart, the number and percentage of DE genes in each category were shown in each sector. h, i Heatmap of h DAM genes and i A1‑specific genes in the TauC/WTC and TauJ/TauC groups. The relative expression level (log ratio) of genes is shown on a scale from red (upregulated) to blue (downregulated). Asterisks indicate significant alterations (p < 0.05) Chang et al. acta neuropathol commun (2021) 9:112 Page 12 of 18 Chang  et al. acta neuropathol commun (2021) 9:112 Page 13 of 18 conditions. Conversely, the impaired energy status of microenvironments (e.g., extracellular space proximal to hippocampal neurons of Tau22 mice (i.e., in an allostatic neuronal soma and synapses), future investigations using situation) causes the activation of AMPK. We hypoth- an in  vivo adenosine sensor [64] will be needed. None- esized that blockade of ENT1 may reduce the entry of theless, the levels of at least three genes (i.e., ADA, CD39, adenosine and, subsequently the cellular level of AMP, CD73) involved in adenosine metabolism were elevated thereby altering the AMP/ATP ratio, and ultimately sup- in the hippocampus of Tau22 mice. A trend of increase pressing AMPK activation in hippocampal neurons of in the transcript level of ADK was also found, but did not Tau22 mice. Our hypothesis is in line with a recent study reach statistical significance. Importantly, J4 treatment demonstrating that genetic deletion of ENT1 in erythro- normalized all these changes (Table  1), suggesting that cytes reduces adenosine uptake and leads to the suppres- adenosine homeostasis was altered in Tau22 mice. It is sion of AMPK [47]. possible that the elevation of ADA in the hippocampus of Accumulating evidence demonstrates that overacti- Tau22 mice may reduce adenosine availability from intra- vation of AMPK in neurons causes synapse loss via an cellular source, while the elevation of CD39 and CD73 autophagy-dependent pathway, and links synaptic integ- increases extracellular adenosine pool, which counteracts rity and energetic failure in neurodegenerative diseases the imbalance of intracellular adenosine level. This may [63]. Here, we found that aberrant AMPK activation was be why the adenosine alteration was not observed in the associated with synaptic loss and reduced basal synaptic hippocampus of Tau22 mice (Additional file  1: Fig. S14). transmission in Tau22 hippocampi (Figs.  1e, 3b, c, 5f, g; J4 treatment reset adenosine homeostasis by blockading Additional file  1: Table  S6). Collectively, J4 suppresses adenosine entry, which results in the decrease of ADA, AMPK overactivation, and normalizes impaired neuronal CD39, and CD73 in the hippocampus of Tau22 mice. plasticity in both APP/PS1 and Tau22 mice (LTP and Collectively, chronic J4 does not induce a major change LTD, respectively; [27]; Fig. 1g, h). in the steady state level of adenosine but rather adenosine Besides the impaired cognitive function, we did not homeostasis. observe any obvious systemic alteration of Tau22 up to An interesting study recently reported that Tau22 mice 12 months except for a slightly lower body weight. Treat- are more susceptible to pentylenetetrazol (PTZ) for sei- ment with J4 did not affect the bodyweight of Tau22 and zure and mortality than WT mice, probably due to the WT mice (Additional file  1: Fig. S12), suggesting that enhanced expression of ADK [65]. This is of great inter - chronic J4 treatment at the condition tested had no obvi- est because J4 is an anti-epileptic agent in a PTZ-induced ous toxicity. Although significant Tau hyperphosphoryla - kindling model [66]. Given that J4 treatment reduced tion and gliosis were observed in the hippocampus of the level of ADK in Tau22 mice (Table  1), it is plausible Tau22 mice of 10–12  months (Figs.  2 and 5), no change that modulation of ADK may contribute to the beneficial in the volume of the whole brain, hippocampus, and ven- effect of J4 in Tau22 mice. No effect of J4 on the expres - tricle were altered (Additional file  1: Fig. S13). Because J4 sions of A adenosine receptor, A adenosine receptor 1 2A is a blocker of ENT1, we measured the levels of adeno- and some of their signaling molecules (i.e., protein kinase sine in the hippocampus of Tau22 mice (11 months old) A, GSK3β, AMPKs) were observed (Additional file  1: but found no difference in either the extracellular or Table S9). the intracellular steady state levels by in  vivo microdi- Although pathogenic Tau is specifically expressed in alysis and tissue extraction coupled to high performance the neurons of mouse models of tauopathy (including liquid chromatography (HPLC), respectively (Figs. S1 Tau22 and rTg4510 mice), reactive microglia and astro- and S14). This is probably because what we measured cytes have been found near neurons that contain high were bulk adenosine concentrations in the extracellu- levels of pathogenic Tau, suggesting that degenerating lar fluid and those inside of cells in the hippocampus. neurons may trigger abnormal gliosis [33] and alter their To assess whether adenosine levels are altered in gene expression profiles [67]. Here, our RNA-seq analysis (See figure on next page.) Fig. 5 Chronic J4 treatment suppresses the activation of microglia and the induction of cytotoxic A1 astrocytes in the hippocampi of Tau22 mice. Mice were treated as indicated ( WTC, black; WTJ, blue; TauC, red; TauJ, green; n = 5–7 in each group) from the age of 3–11 months, and their tissues were subjected to IHC staining and RT‑ qPCR analysis. a CD68 (green) is a marker of reactive microglia and Iba1 (red) is a marker of microglia and the quantification results are shown in (b). c Gene expression of TNF‑α, IL ‑1α, and C1q (C1qa, C1qb, and C1qc) in the hippocampi of treated mice (n = 6–9 in each group) was analyzed, and GAPDH was used as a reference gene. d The intensity of C1q (green) and Iba1 (red) expression was examined and the quantification results are shown in (e). Scale bar, 20 μm. f, g The number of synapses (f) in the hippocampi of treated mice (n = 10–12 in each group) was evaluated by staining with the indicated antibodies (PSD95, green; SYP, red). The quantification results are shown in (g). Scale bar, 5 μm. h Hippocampal sections (20 μm) were prepared and subjected to IHC staining (Lcn2 a marker of reactive astrocytes, green; GFAP a marker of astrocytes, red) and the staining was quantified (i). Scale bar, 20 μm. The data are expressed as the mean ± S.E.M. *p < 0.05 versus the WTC group; p < 0.05 versus the TauC group, one‑ way ANOVA Chang et al. acta neuropathol commun (2021) 9:112 Page 14 of 18 of Tau22 mice showed upregulation of several genes not rescue all inflammation-related pathways as analyzed associated with the disease-associated microglia (DAM, by GO and KEGG (Additional file  1: Fig. S7), 56 of the [55]) in mice and patients with Alzheimer’s Disease 90 dysregulated homeostatic microglia genes (62%; [68]) (Fig.  4h and Additional file  1: Table S7). Although J4 did of Tau22 mice were normalized by J4 treatment (Fig. Chang  et al. acta neuropathol commun (2021) 9:112 Page 15 of 18 S15). These homeostatic microglial genes are commonly Conclusions expressed in microglia of healthy adult brain. Thus, J4 In summary, we provide evidence that an ENT1 inhibi- treatment rescued the dysregulated microglial homeosta- tor (J4) rescues the energy dysfunction (including mito- sis in Tauopathy. chondrial impairment and AMPK overactivation) and Notably, J4 not only ameliorated the energy dysfunc- pathological glial activation, and subsequently improves tion and Tau pathology in neurons (Figs.  1, 2, 3) but synaptic function and memory in tauopathy. Modula- also markedly reduced the activation of CD68-positive tion of adenosine homeostasis by an ENT1 inhibitor (J4) microglia activation and rescued the synapse loss asso- therefore deserves further development in tauopathies ciated with the phagocytic activities of activated micro- and Alzheimer’s Disease. glia (Fig. 5) [34, 57, 69]. Given that CD68 is also a marker of phagocytosis [70], the suppression of CD68 by J4 sug- Abbreviations gest that J4 treatment might decrease the phagocytic Aβ: Amyloid‑beta; ACN: Acetonitrile; ACSF: Artificial cerebrospinal fluid; AD: capacity of microglia and result in the rescue of synapse Alzheimer’s disease; ADA: Adenosine deaminase; ADK: Adenosine kinase; AMPK: AMP‑activated protein kinase; BBB: Blood–brain barrier; BSA: Bovine loss (Fig.  5b), in agreement with the normalization of serum albumin; CBD: Corticobasal degeneration; cDNA: Complementary C1q expression by J4, a microglial complement protein DNA; DAAs: Disease‑associated astrocytes; DAM: Disease ‑associated microglia; known to regulate synaptic phagocytosis by microglia DAVID: Database for Annotation, Visualization and Integrated Discovery; DE: Differentially expressed; DTT: Dithiothreitol; ENT1: Equilibrative nucleoside in a Tau pathology context [34, 69]. We also performed transporter 1; FDR: False discovery rate; fEPSPs: Field excitatory postsynaptic RT-qPCR and established that microglial factors known potentials; FPKM: Fragments per kilobase per million; FTD: Frontotemporal to promote neurotoxic activation of astrocytes were dementia; GO: Gene Ontology; HPβCD: 2‑Hydroxypropyl‑b ‑ cyclodextrin; IAA: Iodoacetamide; IACUC : Institutional Animal Care and Utilization Committee; upregulated in Tau22 mice (C1q and TNFα; Fig.  5c). IVC: In ventilated cages; KEGG: Kyoto Encyclopedia of Genes and Genomes; Consistent with the suppression of microglial activa- LFS: Low‑frequency stimulation; LTD: Long‑term depression; LTP: Long‑term tion, J4 also significantly reduced astrocytic activation, potentiation; MWM: Morris water maze; NDS: Normal donkey serum; NGS: Normal goat serum; PCF: Proteomics Core Facility; PP2A: Protein phosphatase with a particular impact on the A1 signature (Figs.  4i, 2A; PPF: Paired‑pulse facilitation; PSP: Progressive supranuclear palsy; RT ‑ qPCR: 5h, i). We show, for the first time, a pathological acti - Reverse Transcription quantitative PCR; PTZ: Pentylenetetrazol; SBS: Sequenc‑ vation of the A1 neurotoxic astrocytic phenotype in a ing‑by‑synthesis; Tau22: THY ‑ Tau22; TauC: Vehicle‑treated Tau22 mice; TauJ: J4‑treated Tau22; TEAB: Triethylammonium bicarbonate; TFA: Trifluoroacetic tauopathy context using both human and mouse sam- acid; WT: Wild type; WTC : Vehicle‑treated WT mice; WTJ: J4‑treated WT. ples (Additional file  1: Fig. S10 and Table  S8), which may be related to the production of A1-promoting fac- Supplementary Information tors by microglia. Our RNA-seq analysis of Tau22 mice The online version contains supplementary material available at https:// doi. also emphasizes the upregulation of a disease-associ- org/ 10. 1186/ s40478‑ 021‑ 01213‑7. ated astrocytes (DAAs) signature, recently detected in an amyloid Alzheimer’s Disease mouse model (5XFAD) Additional file 1. Supplementary Information. [71] (Additional file  1: Table  S10). These observations are of particular importance since glial activation has Acknowledgements been shown to be instrumental in Alzheimer’s Disease We are grateful to the Proteomics Core Facility of the Institute of Biomedical [5]. In accordance, the beneficial effects of J4 on mem - Sciences, Academia Sinica, and the Neuroscience core facility, Academia Sinica for the LC/MS/MS analysis and the electrophysiology analysis. We thank the ory alterations and neuroplasticity (Figs. 1, 2, 3) is asso- Animal Image Facility, Academia Sinica and Taiwan Animal Consortium for the ciated with the normalization of both A1-promoting technical support in MRI analysis. We acknowledge the UC Davis Alzheimer’s factors and A1 astrocytes in Tau22 mice (Additional Disease Center (USA) for providing human brain specimens from normal, Alzheimer’s Disease, and FTD‑ Tau subjects. We also thank Dr. Peter Davies ( The file  1: Table  S8). Treatment with J4 normalized 60% of Feinstein Institute for Medical Research, New York, USA) for generously provid‑ the elevated DAA genes in Tau22 mice (Additional ing disease‑specific conformational modified tau antibody (MC1). file  1: Table S10), further supporting the beneficial effect Authors’ contributions of J4 on the prevention of abnormal astrocytic activa- CPC conducted behavioral, bioinformatic analysis, and immunostaining of tion. Noteworthy, ENT1 is not only expressed by neu- human brain specimens. YGC performed electrophysiology experiments. rons but also by astrocytes. Earlier studies had reported PYC and TNAN performed biochemical analyses. KCW and FYC performed pharmacokinetics studies. HMC and TNAN managed the animal breeding and that blockade of ENT1 in astrocytes suppresses the genotyping. LWJ and VH provided human brain specimens. KC performed expression of two astrocyte-specific genes (i.e., the Q‑PCR analysis of human brain specimens. LB and DB provided THY ‑ Tau22 type 2 excitatory amino acid transporter (EAAT2) and mice and contributes to bioinformatic analysis. YC, YFL, CJL, and SJC super‑ vised the design of biochemical analyses, electrophysiology experiments, and aquaporin 4 (AQP4)) [72]. Because J4 treatment did not pharmacokinetics studies. CPC, SJC, KC and DB contributed to manuscript affect the level of EAAT2 and AQP4 (Additional file  1: writing. YC oversaw project administration and wrote the final manuscript. All Table  S9), suppression of astrocytic ENT1 by J4 under authors read and approved the final manuscript. the condition tested was unlikely to be significant. Chang et al. acta neuropathol commun (2021) 9:112 Page 16 of 18 Funding 6. Qian W, Shi J, Yin X, Iqbal K, Grundke‑Iqbal I, Gong CX, Liu F (2010) PP2A This research was supported by the Academia Sinica and Ministry of Science regulates tau phosphorylation directly and also indirectly via activat‑ and Technology (MOST 107‑2320‑B‑001‑013‑MY3, AS‑SUMMIT ‑109; MOST ‑ ing GSK‑3beta. J Alzheimers Dis 19:1221–1229. https:// doi. org/ 10. 3233/ 108‑3114‑ Y‑001‑002; AS‑KPQ ‑109‑BioMed). The Neuroscience core facility was jad‑ 2010‑ 1317 supported by the Academia Sinica (AS‑ CFII‑108‑106). The UC Davis Alzheimer’s 7. Sergeant N, Bretteville A, Hamdane M, Caillet‑Boudin ML, Grognet P, Disease Center Biorepository (ADC Biorepository) was supported by the Bombois S, Blum D, Delacourte A, Pasquier F, Vanmechelen E et al (2008) National Institutes of Health (P30‑AG010129). The “Alzheimer & Tauopathies” Biochemistry of Tau in Alzheimer’s disease and related neurological dis‑ laboratory is supported by Inserm, Université Lille, France Alzheimer, programs orders. Expert Rev Proteomics 5:207–224. https:// doi. org/ 10. 1586/ 14789 d’investissements d’avenir LabEx (excellence laboratory) DISTALZ (Develop‑450.5. 2. 207 ment of Innovative Strategies for a Transdisciplinary approach to ALZheimer’s 8. Lee CW, Shih YH, Wu SY, Yang T, Lin C, Kuo YM (2013) Hypoglycemia disease), ANR (ADORASTrAU ANR‑18‑ CE16‑0008 and CoEN 5008), Fondation induces tau hyperphosphorylation. Curr Alzheimer Res 10:298–308. pour la Recherche Médicale, Vaincre Alzheimer, Fondation Plan Alzheimer https:// doi. org/ 10. 2174/ 15672 05011 31003 0009 (ADOMEMOTAU), LilleMétropole Communauté Urbaine, Région Hauts‑ de‑ 9. Merkwirth C, Martinelli P, Korwitz A, Morbin M, Bronneke HS, Jordan SD, France (COGNADORA), and DN2M. Special thanks to PHC Orchid exchange Rugarli EI, Langer T (2012) Loss of prohibitin membrane scaffolds impairs funding for supporting the collaboration between Academia Sinica and mitochondrial architecture and leads to tau hyperphosphorylation and Inserm. neurodegeneration. PLoS Genet 8:e1003021. https:// doi. org/ 10. 1371/ journ al. pgen. 10030 21 Availability of data and materials 10. Carman AJ, Mills JH, Krenz A, Kim DG, Bynoe MS (2011) Adenosine The data that support the findings of this study are available on request to the receptor signaling modulates permeability of the blood–brain barrier. J corresponding author. Neurosci 31:13272–13280. https:// doi. org/ 10. 1523/ JNEUR OSCI. 3337‑ 11. Ethical approval and consent to participate 11. Chen JF, Lee CF, Chern Y (2014) Adenosine receptor neurobiology: All animal studies were conducted following the protocols approved by the overview. Int Rev Neurobiol 119:1–49. https:// doi. org/ 10. 1016/ B978‑0‑ 12‑ Institutional Animal Care and Utilization Committee (IACUC, Academia Sinica, 801022‑ 8. 00001‑5 Taiwan). Human brain samples were obtained from the UC Davis Alzheimer’s 12. Ham J, Evans BA (2012) An emerging role for adenosine and its receptors Disease Center (USA) and the brain banks of Lille, Paris, and Geneva. The IRB in bone homeostasis. Front Endocrinol (Lausanne) 3:113. https:// doi. org/ number of Academic Sinica is AS‑IRB‑BM ‑16013.10. 3389/ fendo. 2012. 00113 13. Lee CF, Chern Y (2014) Adenosine receptors and Huntington’s disease. Int Consent for publication Rev Neurobiol 119:195–232. https:// doi. org/ 10. 1016/ B978‑0‑ 12‑ 801022‑ 8. Not applicable. 00010‑6 14. Anderson CM, Xiong W, Geiger JD, Young JD, Cass CE, Baldwin SA, Parkin‑ Competing interests son FE (1999) Distribution of equilibrative, nitrobenzylthioinosine‑sen‑ Yijuang Chern hold patents on J4 for the treatment of neurodegenerative sitive nucleoside transporters (ENT1) in brain. J Neurochem 73:867–873. diseases.https:// doi. org/ 10. 1046/j. 1471‑ 4159. 1999. 07308 67.x 15. Young JD (2016) The SLC28 (CNT ) and SLC29 (ENT ) nucleoside trans‑ Author details porter families: a 30‑ year collaborative odyssey. Biochem Soc Trans Institute of Biomedical Sciences, Academia Sinica, Nankang, Taipei 115, 44:869–876. https:// doi. org/ 10. 1042/ BST20 160038 Taiwan. School of Pharmacy, National Taiwan University, Taipei, Taiwan. 16. Cunha RA (2016) How does adenosine control neuronal dysfunction and Neuroscience Program of Academia Sinica, Academia Sinica, Taipei, Taiwan. neurodegeneration? J Neurochem 139:1019–1055. https:// doi. org/ 10. Department of Pathology and Laboratory Medicine, University of California 1111/ jnc. 13724 Davis, Sacramento, CA, USA. Univ. Lille, Inserm, CHU Lille, U1172 ‑ LilNCog ‑ 17. Porkka‑Heiskanen T, Kalinchuk AV (2011) Adenosine, energy metabolism Lille Neuroscience & Cognition, 59000 Lille, France. Alzheimer & T auopathies, and sleep homeostasis. Sleep Med Rev 15:123–135. https:// doi. org/ 10. LabEx DISTALZ, LiCEND, 59000 Lille, France. Institute of Cellular and Organis‑1016/j. smrv. 2010. 06. 005 mic Biology, Academia Sinica, Taipei, Taiwan. 18. Alonso‑Andres P, Albasanz JL, Ferrer I, Martin M (2018) Purine ‑related metabolites and their converting enzymes are altered in frontal, parietal Received: 21 April 2021 Accepted: 7 June 2021 and temporal cortex at early stages of Alzheimer’s disease pathology. Brain Pathol 28:933–946. https:// doi. org/ 10. 1111/ bpa. 12592 19. Albasanz JL, Perez S, Barrachina M, Ferrer I, Martin M (2008) Up‑regulation of adenosine receptors in the frontal cortex in Alzheimer’s disease. Brain Pathol 18:211–219. https:// doi. org/ 10. 1111/j. 1750‑ 3639. 2007. 00112.x 20. Temido‑Ferreira M, Ferreira DG, Batalha VL, Marques‑Morgado I, Coelho References JE, Pereira P, Gomes R, Pinto A, Carvalho S, Canas PM et al (2018) Age‑ 1. Cummings J, Blennow K, Johnson K, Keeley M, Bateman RJ, Molinuevo JL, related shift in LTD is dependent on neuronal adenosine A2A receptors Touchon J, Aisen P, Vellas B (2019) Anti‑tau trials for Alzheimer’s disease: a interplay with mGluR5 and NMDA receptors. Mol Psychiatry. https:// doi. report from the EU/US/CTAD task force. J Prev Alzheimers Dis 6:157–163. org/ 10. 1038/ s41380‑ 018‑ 0110‑9 https:// doi. org/ 10. 14283/ jpad. 2019. 14 21. Faivre E, Coelho JE, Zornbach K, Malik E, Baqi Y, Schneider M, Cellai L, 2. Brier MR, Gordon B, Friedrichsen K, McCarthy J, Stern A, Christensen J, Carvalho K, Sebda S, Figeac M et al (2018) Beneficial effect of a selective Owen C, Aldea P, Su Y, Hassenstab J et al (2016) Tau and Abeta imaging, adenosine A2A receptor antagonist in the APPswe/PS1dE9 mouse model CSF measures, and cognition in Alzheimer’s disease. Sci Transl Med of Alzheimer’s disease. Front Mol Neurosci 11:235. https:// doi. org/ 10. 8:338ra366. https:// doi. org/ 10. 1126/ scitr anslm ed. aaf23 62 3389/ fnmol. 2018. 00235 3. Colin M, Dujardin S, Schraen‑Maschke S, Meno ‑ Tetang G, Duyckaerts C, 22. Lee CC, Chang CP, Lin CJ, Lai HL, Kao YH, Cheng SJ, Chen HM, Liao YP, Courade JP, Buee L (2020) From the prion‑like propagation hypothesis Faivre E, Buee L et al (2018) Adenosine augmentation evoked by an ENT1 to therapeutic strategies of anti‑tau immunotherapy. Acta Neuropathol inhibitor improves memory impairment and neuronal plasticity in the 139:3–25. https:// doi. org/ 10. 1007/ s00401‑ 019‑ 02087‑9 APP/PS1 mouse model of Alzheimer’s disease. Mol Neurobiol 55:8936– 4. Ittner LM, Gotz J (2011) Amyloid‑beta and tau—a toxic pas de deux in 8952. https:// doi. org/ 10. 1007/ s12035‑ 018‑ 1030‑z Alzheimer’s disease. Nat Rev Neurosci 12:65–72. https:// doi. org/ 10. 1038/ 23. Silva AC, Lemos C, Goncalves FQ, Pliassova AV, Machado NJ, Silva HB, nrn29 67 Canas PM, Cunha RA, Lopes JP, Agostinho P (2018) Blockade of adenosine 5. Laurent C, Buée L, Blum D (2018) Tau and neuroinflammation: what A2A receptors recovers early deficits of memory and plasticity in the impact for Alzheimer’s disease and tauopathies? Biomed J 41:21–33. https:// doi. org/ 10. 1016/j. bj. 2018. 01. 003 Chang  et al. acta neuropathol commun (2021) 9:112 Page 17 of 18 triple transgenic mouse model of Alzheimer’s disease. Neurobiol Dis 40. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, 117:72–81. https:// doi. org/ 10. 1016/j. nbd. 2018. 05. 024 Salzberg SL, Rinn JL, Pachter L (2012) Differential gene and transcript 24. Viana da Silva S, Haberl MG, Zhang P, Bethge P, Lemos C, Goncalves N, expression analysis of RNA‑seq experiments with TopHat and Cufflinks. Gorlewicz A, Malezieux M, Goncalves FQ, Grosjean N et al (2016) Early Nat Protoc 7:562–578. https:// doi. org/ 10. 1038/ nprot. 2012. 016 synaptic deficits in the APP/PS1 mouse model of Alzheimer’s disease 41. Nolte H, MacVicar TD, Tellkamp F, Kruger M (2018) Instant clue: a software involve neuronal adenosine A2A receptors. Nat Commun 7:11915. suite for interactive data visualization and analysis. Sci Rep 8:12648. https:// doi. org/ 10. 1038/ ncomm s11915https:// doi. org/ 10. 1038/ s41598‑ 018‑ 31154‑6 25. Fang EF, Hou Y, Palikaras K, Adriaanse BA, Kerr JS, Yang B, Lautrup S, 42. da Huang W, Sherman BT, Lempicki RA (2009) Bioinformatics enrichment Hasan‑ Olive MM, Caponio D, Dan X et al (2019) Mitophagy inhibits tools: paths toward the comprehensive functional analysis of large gene amyloid‑β and tau pathology and reverses cognitive deficits in models lists. Nucleic Acids Res 37:1–13. https:// doi. org/ 10. 1093/ nar/ gkn923 of Alzheimer’s disease. Nat Neurosci 22:401–412. https:// doi. org/ 10. 1038/ 43. da Huang W, Sherman BT, Lempicki RA (2009) Systematic and integrative s41593‑ 018‑ 0332‑9 analysis of large gene lists using DAVID bioinformatics resources. Nat 26. Lauretti E, Li JG, Di Meco A, Praticò D (2017) Glucose deficit triggers tau Protoc 4:44–57. https:// doi. org/ 10. 1038/ nprot. 2008. 211 pathology and synaptic dysfunction in a tauopathy mouse model. Transl 44. Jicha GA, Bowser R, Kazam IG, Davies P (1997) Alz‑50 and MC‑1, a new Psychiatry 7:e1020. https:// doi. org/ 10. 1038/ tp. 2016. 296 monoclonal antibody raised to paired helical filaments, recognize confor ‑ 27. Ma T, Chen Y, Vingtdeux V, Zhao H, Viollet B, Marambaud P, Klann E mational epitopes on recombinant tau. J Neurosci Res 48:128–132 (2014) Inhibition of AMP‑activated protein kinase signaling alleviates 45. Mellone M, Kestoras D, Andrews MR, Dassie E, Crowther RA, Stokin GB, impairments in hippocampal synaptic plasticity induced by amyloid β. Tinsley J, Horne G, Goedert M, Tolkovsky AM et al (2013) Tau pathology J Neurosci 34:12230–12238. https:// doi. org/ 10. 1523/ jneur osci. 1694‑ 14. is present in vivo and develops in vitro in sensory neurons from human 2014 P301S tau transgenic mice: a system for screening drugs against tauopa‑ 28. Vingtdeux V, Davies P, Dickson DW, Marambaud P (2011) AMPK is abnor‑ thies. J Neurosci 33:18175–18189. https:// doi. org/ 10. 1523/ JNEUR OSCI. mally activated in tangle‑ and pre ‑tangle ‑bearing neurons in Alzheimer’s 4933‑ 12. 2013 disease and other tauopathies. Acta Neuropathol 121:337–349. https:// 46. Martin L, Latypova X, Wilson CM, Magnaudeix A, Perrin ML, Terro F (2013) doi. org/ 10. 1007/ s00401‑ 010‑ 0759‑x Tau protein phosphatases in Alzheimer’s disease: the leading role of 29. Herzig S, Shaw RJ (2018) AMPK: guardian of metabolism and mitochon‑ PP2A. Ageing Res Rev 12:39–49. https:// doi. org/ 10. 1016/j. arr. 2012. 06. 008 drial homeostasis. Nat Rev Mol Cell Biol 19:121–135. https:// doi. org/ 10. 47. Sayama S, Song A, Brown BC, Couturier J, Cai X, Xu P, Chen C, Zheng Y, 1038/ nrm. 2017. 95 Iriyama T, Sibai B et al (2020) Maternal erythrocyte ENT1‑mediated AMPK 30. Turnley AM, Stapleton D, Mann RJ, Witters LA, Kemp BE, Bartlett PF (1999) activation counteracts placental hypoxia and supports fetal growth. JCI Cellular distribution and developmental expression of AMP‑activated Insight. https:// doi. org/ 10. 1172/ jci. insig ht. 130205 protein kinase isoforms in mouse central nervous system. J Neurochem 48. Domise M, Didier S, Marinangeli C, Zhao H, Chandakkar P, Buee L, Viollet 72:1707–1716. https:// doi. org/ 10. 1046/j. 1471‑ 4159. 1999. 721707.x B, Davies P, Marambaud P, Vingtdeux V (2016) AMP‑activated protein 31. Medina‑Pulido L, Molina‑Arcas M, Justicia C, Soriano E, Burgaya F, Planas kinase modulates tau phosphorylation and tau pathology in vivo. Sci Rep AM, Pastor‑Anglada M (2013) Hypoxia and P1 receptor activation regulate 6:26758. https:// doi. org/ 10. 1038/ srep2 6758 the high‑affinity concentrative adenosine transporter CNT2 in differenti‑ 49. Horman S, Browne G, Krause U, Patel J, Vertommen D, Bertrand L, ated neuronal PC12 cells. Biochem J 454:437–445. https:// doi. org/ 10. Lavoinne A, Hue L, Proud C, Rider M (2002) Activation of AMP‑activated 1042/ BJ201 30231 protein kinase leads to the phosphorylation of elongation factor 2 and an 32. Pastor‑Anglada M, Errasti‑Murugarren E, Aymerich I, Casado FJ (2007) inhibition of protein synthesis. Curr Biol 12:1419–1423. https:// doi. org/ 10. Concentrative nucleoside transporters (CNTs) in epithelia: from absorp‑1016/ s0960‑ 9822(02) 01077‑1 tion to cell signaling. J Physiol Biochem 63:97–110. https:// doi. org/ 10. 50. Li S, Xiong GJ, Huang N, Sheng ZH (2020) The cross‑talk of energy sensing 1007/ BF031 74089 and mitochondrial anchoring sustains synaptic efficacy by maintaining 33. Schindowski K, Bretteville A, Leroy K, Begard S, Brion JP, Hamdane M, Buee presynaptic metabolism. Nat Metab 2:1077–1095. https:// doi. org/ 10. L (2006) Alzheimer’s disease‑like tau neuropathology leads to memory 1038/ s42255‑ 020‑ 00289‑0 deficits and loss of functional synapses in a novel mutated tau transgenic 51. Woods A, Dickerson K, Heath R, Hong SP, Momcilovic M, Johnstone SR, 2+ mouse without any motor deficits. Am J Pathol 169:599–616. https:// doi. Carlson M, Carling D (2005) Ca /calmodulin‑ dependent protein kinase org/ 10. 2353/ ajpath. 2006. 060002 kinase‑beta acts upstream of AMP ‑activated protein kinase in mamma‑ 34. Carvalho K, Faivre E, Pietrowski MJ, Marques X, Gomez‑Murcia V, Deleau lian cells. Cell Metab 2:21–33. https:// doi. org/ 10. 1016/j. cmet. 2005. 06. 005 A, Huin V, Hansen JN, Kozlov S, Danis C et al (2019) Exacerbation of C1q 52. DuBoff B, Gotz J, Feany MB (2012) Tau promotes neurodegeneration via dysregulation, synaptic loss and memory deficits in tau pathology linked DRP1 mislocalization in vivo. Neuron 75:618–632. https:// doi. org/ 10. to neuronal adenosine A2A receptor. Brain 142:3636–3654. https:// doi. 1016/j. neuron. 2012. 06. 026 org/ 10. 1093/ brain/ awz288 53. Cha MY, Cho HJ, Kim C, Jung YO, Kang MJ, Murray ME, Hong HS, Choi 35. Huin V, Deramecourt V, Caparros‑Lefebvre D, Maurage CA, Duyckaerts C, YJ, Choi H, Kim DK et al (2015) Mitochondrial ATP synthase activity is Kovari E, Pasquier F, Buée‑Scherrer V, Labreuche J, Behal H et al (2016) The impaired by suppressed O‑ GlcNAcylation in Alzheimer’s disease. Hum MAPT gene is differentially methylated in the progressive supranuclear Mol Genet 24:6492–6504. https:// doi. org/ 10. 1093/ hmg/ ddv358 palsy brain. Mov Disord 31:1883–1890. https:// doi. org/ 10. 1002/ mds. 54. Zhang Y, Chen K, Sloan SA, Bennett ML, Scholze AR, O’Keeffe S, Phatnani 26820 HP, Guarnieri P, Caneda C, Ruderisch N et al (2014) An RNA‑sequencing 36. Chang CP, Lee CT, Hou WH, Lin MS, Lai HL, Chien CL, Chang C, Cheng transcriptome and splicing database of glia, neurons, and vascular cells of PL, Lien CC, Chern Y (2016) Type VI adenylyl cyclase negatively regulates the cerebral cortex. J Neurosci 34:11929–11947. https:// doi. org/ 10. 1523/ GluN2B‑mediated LTD and spatial reversal learning. Sci Rep 6:22529. JNEUR OSCI. 1860‑ 14. 2014 https:// doi. org/ 10. 1038/ srep2 2529 55. Keren‑Shaul H, Spinrad A, Weiner A, Matcovitch‑Natan O, Dvir ‑Szternfeld 37. Van der Jeugd A, Ahmed T, Burnouf S, Belarbi K, Hamdame M, Grosjean R, Ulland TK, David E, Baruch K, Lara‑Astaiso D, Toth B et al (2017) A ME, Humez S, Balschun D, Blum D, Buee L et al (2011) Hippocampal unique microglia type associated with restricting development of Alzhei‑ tauopathy in tau transgenic mice coincides with impaired hippocampus‑ mer’s disease. Cell 169(1276–1290):e1217. https:// doi. org/ 10. 1016/j. cell. dependent learning and memory, and attenuated late‑phase long‑term 2017. 05. 018 depression of synaptic transmission. Neurobiol Learn Mem 95:296–304. 56. Liddelow SA, Guttenplan KA, Clarke LE, Bennett FC, Bohlen CJ, Schirmer https:// doi. org/ 10. 1016/j. nlm. 2010. 12. 005 L, Bennett ML, Munch AE, Chung WS, Peterson TC et al (2017) Neuro‑ 38. Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: a flexible trimmer for toxic reactive astrocytes are induced by activated microglia. Nature Illumina sequence data. Bioinformatics 30:2114–2120. https:// doi. org/ 10. 541:481–487. https:// doi. org/ 10. 1038/ natur e21029 1093/ bioin forma tics/ btu170 57. Gratuze M, Leyns CE, Sauerbeck AD, St‑Pierre MK, Xiong M, Kim N, Remo ‑ 39. Kim D, Langmead B, Salzberg SL (2015) HISAT: a fast spliced aligner with lina Serrano J, Tremblay ME, Kummer TT, Colonna M et al (2020) Impact of low memory requirements. Nat Methods 12:357–360. https:// doi. org/ 10. TREM2R47H variant on tau pathology‑induced gliosis and neurodegen‑ 1038/ nmeth. 3317 eration. J Clin Invest. https:// doi. org/ 10. 1172/ JCI13 8179 Chang et al. acta neuropathol commun (2021) 9:112 Page 18 of 18 58. Chatterjee S, Cassel R, Schneider‑Anthony A, Merienne K, Cosquer B, 67. Wang H, Li Y, Ryder JW, Hole JT, Ebert PJ, Airey DC, Qian HR, Logsdon B, Tzeplaeff L, Halder Sinha S, Kumar M, Chaturbedy P, Eswaramoorthy M Fisher A, Ahmed Z et al (2018) Genome‑ wide RNAseq study of the molec‑ et al (2018) Reinstating plasticity and memory in a tauopathy mouse ular mechanisms underlying microglia activation in response to patho‑ model with an acetyltransferase activator. EMBO Mol Med. https:// doi. logical tau perturbation in the rTg4510 tau transgenic animal model. Mol org/ 10. 15252/ emmm. 20170 8587 Neurodegener 13:65. https:// doi. org/ 10. 1186/ s13024‑ 018‑ 0296‑y 59. Clarke LE, Liddelow SA, Chakraborty C, Munch AE, Heiman M, Barres BA 68. Li Q, Cheng Z, Zhou L, Darmanis S, Neff NF, Okamoto J, Gulati G, Bennett (2018) Normal aging induces A1‑like astrocyte reactivity. Proc Natl Acad ML, Sun LO, Clarke LE et al (2019) Developmental heterogeneity of micro‑ Sci U S A 115:E1896–E1905. https:// doi. org/ 10. 1073/ pnas. 18001 65115 glia and brain myeloid cells revealed by deep single‑ cell RNA sequenc‑ 60. Bi F, Huang C, Tong J, Qiu G, Huang B, Wu Q, Li F, Xu Z, Bowser R, Xia XG ing. Neuron 101:207‑223.e210. https:// doi. org/ 10. 1016/j. neuron. 2018. 12. et al (2013) Reactive astrocytes secrete lcn2 to promote neuron death. 006 Proc Natl Acad Sci U S A 110:4069–4074. https:// doi. org/ 10. 1073/ pnas. 69. Dejanovic B, Huntley MA, De Mazière A, Meilandt WJ, Wu T, Srinivasan K, 12184 97110 Jiang Z, Gandham V, Friedman BA, Ngu H et al (2018) Changes in the syn‑ 61. Kerr JS, Adriaanse BA, Greig NH, Mattson MP, Cader MZ, Bohr VA, Fang EF aptic proteome in tauopathy and rescue of tau‑induced synapse loss by (2017) Mitophagy and Alzheimer’s disease: cellular and molecular mecha‑ C1q antibodies. Neuron 100:1322‑1336.e1327. https:// doi. org/ 10. 1016/j. nisms. Trends Neurosci 40:151–166. https:// doi. org/ 10. 1016/j. tins. 2017. 01. neuron. 2018. 10. 014 002 70. Gunner G, Cheadle L, Johnson KM, Ayata P, Badimon A, Mondo E, Nagy 62. Zheng J, Akbari M, Schirmer C, Reynaert ML, Loyens A, Lefebvre B, Buee L, MA, Liu L, Bemiller SM, Kim KW et al (2019) Sensory lesioning induces Croteau DL, Galas MC, Bohr VA (2020) Hippocampal tau oligomerization microglial synapse elimination via ADAM10 and fractalkine signaling. Nat early in tau pathology coincides with a transient alteration of mitochon‑ Neurosci 22:1075–1088. https:// doi. org/ 10. 1038/ s41593‑ 019‑ 0419‑y drial homeostasis and DNA repair in a mouse model of tauopathy. Acta 71. Habib N, McCabe C, Medina S, Varshavsky M, Kitsberg D, Dvir‑Szternfeld R, Neuropathol Commun 8:25. https:// doi. org/ 10. 1186/ s40478‑ 020‑ 00896‑8 Green G, Dionne D, Nguyen L, Marshall JL et al (2020) Disease‑associated 63. Domise M, Sauve F, Didier S, Caillerez R, Begard S, Carrier S, Colin M, astrocytes in Alzheimer’s disease and aging. Nat Neurosci 23:701–706. Marinangeli C, Buee L, Vingtdeux V (2019) Neuronal AMP‑activated https:// doi. org/ 10. 1038/ s41593‑ 020‑ 0624‑8 protein kinase hyper‑activation induces synaptic loss by an autophagy‑ 72. Lee MR, Ruby CL, Hinton DJ, Choi S, Adams CA, Young Kang N, Choi DS mediated process. Cell Death Dis 10:221. https:// doi. org/ 10. 1038/ (2013) Striatal adenosine signaling regulates EAAT2 and astrocytic AQP4 s41419‑ 019‑ 1464‑x expression and alcohol drinking in mice. Neuropsychopharmacology 64. Peng W, Wu Z, Song K, Zhang S, Li Y, Xu M (2020) Regulation of sleep 38:437–445. https:// doi. org/ 10. 1038/ npp. 2012. 198 homeostasis mediator adenosine by basal forebrain glutamatergic neu‑ rons. Science. https:// doi. org/ 10. 1126/ scien ce. abb05 56 Publisher’s Note 65. Gomez‑Murcia V, Sandau U, Ferry B, Parrot S, Laurent C, Basquin M, Buee Springer Nature remains neutral with regard to jurisdictional claims in pub‑ L, Boison D, Blum D (2020) Hyperexcitability and seizures in the THY‑ lished maps and institutional affiliations. Tau22 mouse model of tauopathy. Neurobiol Aging 94:265–270. https:// doi. org/ 10. 1016/j. neuro biola ging. 2020. 06. 004 66. Ho SY, Chen IC, Chang KC, Lin HR, Tsai CW, Lin CJ, Liou HH (2020) Equili‑ brative nucleoside transporters‑1 inhibitors act as anti‑ epileptic agents by inhibiting glutamatergic transmission. Front Neurosci 14:610898. https:// doi. org/ 10. 3389/ fnins. 2020. 610898 Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Acta Neuropathologica CommunicationsSpringer Journals

Published: Jun 22, 2021

Keywords: Alzheimer’s disease; Tauopathy; Adenosine; AMPK; ENT1

References