HMGCS2-Induced Autophagic Degradation of Tau Involves Ketone Body and ANKRD24

Abstract

Background:

Accumulation of hyperphosphorylated Tau (pTau) contributes to the formation of neurofibrillary tangles in Alzheimer’s disease (AD), and targeting Tau/pTau metabolism has emerged as a therapeutic approach. We have previously reported that mitochondrial 3-hydroxy-3-methylglutaryl-COA synthase 2 (HMGCS2) is involved in AD by promoting autophagic clearance of amyloid-β protein precursor via ketone body-associated mechanism, whether HMGCS2 may also regulate Tau metabolism remains elusive.

Objective:

The present study was to investigate the role of HMGCS2 in Tau/p degradation.

Methods:

The protein levels of Tau and pTau including pT217 and pT181, as well as autophagic markers LAMP1 and LC3-II were assessed by western blotting. The differentially regulated genes by HMGCS2 were analyzed by RNA sequencing. Autophagosomes were assessed by transmission electron microscopy.

Results:

HMGCS2 significantly decreased Tau/pTau levels, which was paralleled by enhanced formation of autophagic vacuoles and prevented by autophagic regulators chloroquine, bafilomycin A1, 3-methyladenine, and rapamycin. Moreover, HMGCS2-induced alterations of LAMP1/LC3-II and Tau/pTau levels were mimicked by ketone body acetoacetate or β-hydroxybutyrate. Further RNA-sequencing identified ankyrin repeat domain 24 (ANKRD24) as a target gene of HMGCS2, and silencing of ANKRD24 reduced LAMP1/LC3-II levels, which was accompanied by the altered formation of autophagic vacuoles, and diminished the effect of HMGCS2 on Tau/pTau.

Conclusion:

HMGCS2 promoted autophagic clearance of Tau/pTau, in which ketone body and ANKRD24 played an important role.

Keywords

Alzheimer’s disease ANKRD24 autophagy HMGCS2 ketone body Tau

INTRODUCTION

Alzheimer’s disease (AD) is an age-related neurodegenerative disease characterized by memory loss and cognitive dysfunction [1]. The pathological hallmarks of AD include amyloid-β (Aβ) peptides-derived senile plaques and neurofibrillary tangles (NFTs) formed by hyperphosphorylated and aggregated Tau [2, 3]. Although the etiology of AD remains inconclusive, new lines of evidence favor both Aβ- and Tau-centered hypothesis [4, 5]. It is likely that Aβ and Tau work together and promote healthy neurons into disease state, and toxic Tau could enhance Aβ toxicity through feedback loop [6].

The hyperphosphorylated Tau (pTau) not only impairs microtubule stability, but also interferes with normal nuclear-cytoplasmic transport [7, 8], contributing to the formation of NFTs, neurodegeneration and cognitive dysfunctions [5, 9]. Among the critical pTau sites, pT231 and pS262 convert Tau into an inhibitory molecule that sequesters normal microtubule-associated proteins from microtubules [10], whereas pS396 promotes self-aggregation of Tau into filaments [11]. It is recently reported that plasma pT217 and pT181 could be used for screening individuals with underlying AD pathology and monitor disease progression [12, 13]. Importantly, inhibition of Tau phosphorylation attenuates neuronal cell death [14], and reduction of Tau rescues Aβ-induced neurotoxicity in primary neurons of mice [15, 16]. Thus, targeting Tau/pTau represents a therapeutic approach for AD [17 –20].

Mitochondrial 3-hydroxy-3-methylglutaryl-COA synthase 2 (HMGCS2, EC 2.3.3.10) is a key component of HMG-CoA cycle, by which the ketone bodies (KB) acetoacetate (AcAc) and β-hydroxybutyrate (BHB), in addition to nicotinamide adenine dinucleotide are produced [21]. In the brain, loss of HMGCS2 does not seem to affect brain development in some mammal species including dolphins and elephants [22]. However, our previous reports demonstrate that HMGCS2 expression is reduced in aged APP/PS1 mice, an animal model of AD [23], and HMGCS2 promotes autophagic clearance of amyloid-β protein precursor (AβPP) via KB-associated mechanism [24]. In line with these findings, oral supplement of KB significantly improves cognitive function in AD patients, and ketone ester diet decreases amyloid and Tau pathologies in the mouse model of AD [26]. Thus, KB relative to HMGCS2 is beneficial for AD [27], whereas the potential mechanisms remain incompletely understood.

In this study, we investigated the role of HMGCS2 in Tau metabolism and found that HMGCS2 potently promoted autophagic clearance of all Tau forms tested. We further defined that KB and ankyrin repeat domain 24 (ANKRD24) were involved in this regulation.

MATERIALS AND METHODS

Antibodies and reagents

Information for antibodies was as follows: HMGCS2 (Cat#: ab137043, RRID: AB_2749817; 1:10000), non-phosphorylated Tau (non-pTau, recognizes non-phosphorylated Tau except pS262, Cat#: ab64193, RRID:AB_1143333; 1:1000), Tau-5 (recognizes all forms of Tau, Cat#: ab80579, RRID:AB_1603723; 1:1000), pS396 Tau (Cat#: ab109390, RRID: AB_10860822; 1:1000), pS262 Tau (Cat#: ab131354, RRID:AB_11156689; 1:1000), pT231 Tau (Cat#: ab151559, RRID: AB_2893278;1:1000), pT217 Tau (Cat#: ab75775, RRID: AB_2141775; 1:1000), pT181 Tau (Cat#: ab254409, RRID: AB_2905609; 1:1000), CAMKII (Cat#: ab52476, RRID:AB_868641; 1:1000), PP2A (Cat#: ab32141, RRID: AB_2169649; 1:3000), LC3-II (Cat#: ab48394, RRID:AB_881433; 1:2000), AKT (Cat#: 10176-2-AP, RRID:AB_2224574; 1:1000), phospho (Ser473)-AKT (Cat#: 66444-1-lg, RRID:AB_2782958; 1:1000), CDK5 (Cat#: 10430-1-P, RRID:AB_2078859; 1:1000), GSK3β (Cat#: 22104-1-AP, RRID:AB_2878997; 1:1000), phosphor (Ser389)-GSK3β(Cat#: 14850-1-AP, RRID:AB_2878085; 1:1000), IDE (Cat#:21728-1-AP, RRI-D:AB_10804773; 1:1000), Neprilysin (Cat#: 18008-1-AP, RRID:AB_214654; 1:1000), TTBK2 (Cat#: 15072-1-AP, RRID:AB_2211507; 1:1000), LAMP1 (Cat#: 21997-1-AP, RRID:AB_2878966; 1:1000), AMPK (Cat#: 5831, RRID:AB_10622186; 1:1000), phospho (Thr172)-AMPK (Cat#: 2535, RRID:AB_331250; 1:1000), ANKRD24 (Cat#: sc-241811, RRID:AB_10851440;1:1000), the primary antibody against GAPDH (Cat#: 60004-1-lg, RRID:AB_2107436; 1:10000), the horseradish peroxidase-conjugated anti-mouse (Cat#: SA00001-1, RRID:AB_2722565; 1:5000), and anti-rabbit secondary antibodies (Cat#: SA00001-2, RRID:AB_2722564; 1:5000).

Information for reagents was as follows: Mg132 (Cat#: M7449), Bafilomycin A1 (BafA1, Cat#: 19-148), 3-Methyladenine (3-MA, Cat#: M9281), AcAc (Cat#: A8509), and BHB (Cat#: H6501) were purchased from Sigma-Aldrich (St. Louis, MO); Rapamycin (Rapa, Cat#: AY-22989) and Chloroquine (CQ, Cat#: NSC4375) was purchased from Selleckchem (Houston, TX). Mg132, BafA1, and Rapa were dissolved in dimethylsulphoxide (DMSO, Dingguo, Beijing China, Cat#: DH105-2). The work solution of DMSO was no more than 0.1% (vol/vol). AcAc and BHB were dissolved in phosphate buffer (10 mM, PH: 7.5); 3-MA and CQ were dissolved in sterile water. Dilutions were performed asrequired.

Cell culture and pharmacologic treatments

U251MG (iCell Cat#: h219, RRID: CVCL_0021) human astrocytoma cell line was cultured in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, Cat#:11965092) supplemented with 10% fetal bovine serum (FBS, Hyclone, Cat#: SH30080.03), 100 U/ml penicillin G sodium, and 100μg/ml streptomycin sulfate. SH-SY5Y (The European Collection of Authenticated Cell Cultures, Cat#:94030304, RRID: CVCL_0019) human neuroblastoma cell line was cultured in DMEM/F-12 (Gibco, Cat#: 11320033) supplemented with 10% FBS (Gibco, Cat#:16140071), 100 mg/ml streptomycin sulfate, and 100 U/ml penicillin G sodium. HEK293 (American Type Culture Collection, Cat#: PTA-4488, RRID: CVCL_0045) human embryonic kidney 293 cell line stably expressing human full-length APP695 was maintained in DMEM (Gibco, Cat#: 11965092) with 10% FBS (Hyclone, Cat#:SH30080.03) containing 0.2 mg/ml G418 (Invitrogen, Cat#: 10131035). All cell lines were cultured at 37°C in a humidified 5% CO_ 2 atmosphere. These cell lines are not listed in the International Cell Line Authentication Committee database of cross-contaminated or misidentified cell lines. All the cells used in the study were no more than twenty passages.

Unless otherwise stated, the concentrations and actuation duration parameters of pharmacologic treatments for cells were designated as follows: Rapa 2μM for 24 h, Mg132 1μM for 24 h, 3-MA 10 mM for 24 h, BafA1 200 nM for 24 h, AcAc 5 mM for 24 h, BHB 4 mM for 24 h, and CQ 100μM for 4 h. The protocols employed here were based on published reports to avoid cell toxicity [24 , 28– 31].

Plasmid and small interfering RNA (siRNA) transfection

HMGCS2 (Cat#:SC116704) and the control vector PCMV6 (Cat#: PCMV6XL5) were purchased from Origene (Rockville, MD). The HMGCS2 and the ANKRD24-specific siRNA were purchased from Genepharma (Shanghai, China). The selected interfering sequence (No. #2, Supplementary Figure 1) for HMGCS2: sense: 5’-GAGGAGUUCACAGAAAUAATT-3’ and antisense: 5’-UUAUUUCUGUGAACUCCUCTT-3’; The selected interfering sequence (No. &2, Supplementary Figure 4) for ANKRD24: sense: 5’-GCUCCAGAAACCAAAGUUATT-3’ and antisense: 5’-UAACUUUGGUUUCUGGAGCTT-3’; The interfering sequence for non-targeting siRNA: sense: 5’-UUCUCCGAACGUGUCACGUTT-3’ and antisense: 5’-ACGUGACACGUUCGGAGAATT-3’, which was used as the negative control (siControl).

For plasmid transfection, 2.5μg HMGCS2 or Vector plasmid was mixed with 5μl lipofectamine^®3000 reagent (Cat#: L3000015, Invitrogen, Carlsbad, CA, USA) plus 5μl p3000trademark reagent (Cat#: L3000075, Invitrogen, Carlsbad, CA, USA) in 250μl Opti-MEM Reduces Serum Media (Cat#:31985088, Gibco). For siRNA transfection, 30 pmol HMGCS2 siRNA, ANKRD24 siRNA, or siControl was mixed with 7.5μl Lipofectaminetrademark RNAiMAX reagent (Cat#: 13778075, Invitrogen, Carlsbad, CA, USA) in 250μl Opti-MEM Reduces Serum Media. For plasmid and siRNA cotransfection, 1μg HMGCS2 and 50 pmol ANKRD24 siRNA were mixed with 5μl Lipofectaminetrademark 2000 reagent (Cat#:1168019, Invitrogen, Carlsbad, CA, USA) in the 250μl Opti-MEM Reduces Serum Media. For double siRNA cotransfection, 25 pmol HMGCS2 siRNA and 25 pmol ANKRD24 siRNA were mixed with 7.5μl Lipofectaminetrademark RNAiMAX reagent in 250μl Opti-MEM Reduces Serum Media. Cells were seeded in 6-well plates (Corning Life sciences, Tewksbury, MA) at a density of 5×10⁵ cells/well and harvested 48 h after transfection. The transfection activity of the interested gene was evaluated by western blotting analysis.

Western blotting analysis

Cellular samples were homogenized in ice-cold RIPA buffer (50 mM Tris, 1 mM EDTA, 150 mM NaCl, 1% Triton X-100, 0.1% SDS, 0.5% sodium deoxycholate, Cat#: P0013B, Beyotime, Haimen, China) supplemented with protease inhibitors (Cat#: WB-0181, Beyotime, Haimen, China) and phosphatase inhibitors mixture (Cat#: P1081, Beyotime, Haimen, China). Samples were sonicated on ice and centrifuged at 10,000 g for 10 min at 4°C. Protein concentrations were measured by a BCA Protein Assay Kit (Cat#: P0011, Beyotime, Haimen, China). A total of 10– 50μg of proteins were sampled and then separated on 8% SDS-PAGE, then transferred onto 0.45μM PVDF membranes (Cat#: IPVH00010, Merck Millipore, Billerica, MA). The membranes were blocked for 1 h in Tris-buffered saline containing 0.05% Tween-20 and 5% non-fat milk, and were probed with the respective primary antibodies against the target protein overnight at 4°C. The blots were then washed and incubated for 1 h at room temperature with horseradish peroxidase-conjugated anti-rabbit or anti-mouse secondary antibodies. Bands were visualized by enhanced chemi-luminescence reagent (Cat#: P0018M, Beyotime, Haimen, China) and Fusion FX5 image analysis system (Vilber Lourmat, Marne-la-Vall $\overset{´}{e}$ e, France). Relative protein intensities were calculated by Quantity One 1-D Analysis Software (RRID: SCR_014280, Bio-Rad, Hercules, CA) normalized to GAPDH. No membrane was stripped andre-probed.

RNA sequencing and bioinformation analysis

Total RNA from SH-SY5Y cells transfected with HMGCS2 or the corresponding Vector was extracted using RNAiso plus (Cat#: 9109, Takara, Dalian, Liaoning, China) according to the manufacturer’s instructions. RNA quality was determined using an Agilent 2100 bioanalyzer (Agilent 2100 Bioanalyzer Instrument, RRID: SCR_019389), Stranded RNA sequencing libraries were constructed from 2μg of total RNAs using the KC-DigitalTM stranded mRNA Library Prep Kit for Illumina^® (Cat#: DR08502, Wuhan Seqhealth Co., Ltd. China). This kit eliminates duplication bias in PCR and sequencing steps, by using a unique molecular identifier of eight random bases to label the pre-amplified cDNA molecules. The library products were enriched, quantified, and deeply sequenced on a Hiseq X10 sequencer (Illumina, RRID: SCR_010233). Raw sequencing data were first filtered by Trimmomatic (version 0.36, RRID: SCR_011848), low-quality reads were discarded, and the reads contaminated with adaptor sequences were trimmed. Clean Reads were further treated with in house scripts to eliminate duplication bias introduced in library preparation and sequencing. Differentially expressed genes (DEGs) were identified by DEseq2 with the criteria of absolute log2 (fold change)≥1 and FDR corrected p-value≤0.05. Biological function analysis of DEGs was enriched by Gene ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway with a corrected p-value cut-off of 0.05 to judge statistically significant enrichment. The protein-protein interaction network analysis of DEGs was constructed and visualized by using the biological online databases (IntAct and BioGRID). The RNA sequencing dataset is available in the SRA repository (https://dataview.ncbi.nlm.nih.gov/object/PRJNA834033).

Transmission electron microscopy (TEM)

For electron microscopy, cells were prefixed with a 3% glutaraldehyde for 2 h at room temperature, and then fixed in 1% osmium tetroxide for 2 h at 4°C. After dehydration in an ethanol gradient (50% for 15 min, 70% overnight, 80% and 90% in each for 15 min, and 100% for 20 min two times), the cells were incubated with propanone and embedded in Epon812 (Cat#: 45345, Sima-Aldrich, St. Louis, MO), the ultrathin sections were stained with uranyl acetate and lead citrate. The stained sections were examined with JEM-1400-FLASH Transmission Electron Microscope (JEOL, RRID: SCR_020179) at 80 kV.

Enzyme-linked immunosorbent assay for Aβ₄₀ and Aβ₄₂

Human Aβ₄₀ and Aβ₄₂ in the cell lysates and medium of HEK-APP were measured by the Enzyme-linked immunosorbent assay kits (Elabscience, Cat#: E-EL-HO543c for Aβ₄₂; Cat#: E-EL-H0542c for Aβ₄₀) 48 h post-transfection with HMGCS2, according to the manufacturer’s guidelines.

Statistical analysis

Data are presented as mean±standard error of the mean (SEM) from at least three independent experiments and were analyzed using SPSS 22.0 software (RRID: SCR_019096, Chicago, IL, USA) by two-tailed independent samples Student’s t test between two groups or one-way analysis of variance (ANOVA) with a least-significant difference (LSD) multiple comparison test among multiple groups after variances equality assessed with Levene’s test. Differences were considered significant when p < 0.05. No statistical test for outliers was conducted and no data point were excluded.

RESULTS

HMGCS2 reduced the protein levels of different forms of Tau protein

Metabolic cooperation between astrocytes and neurons is critical for brain function [32]. Importantly, abnormally phosphorylated Tau is frequently found in astrocytes, in addition to neurons [33]. To determine whether HMGCS2 modulates Tau metabolism in both cell types, we assessed the protein levels of non-pTau, total Tau (Tau5), and pTau including pS396, pS262, pT231, in particular pT217 and pT181, two novel promising biomarkers for AD [34, 35], in astrocytic U251MG and neuronal SH-SY5Y cells transfected with HMGCS2. As shown in Fig. 1a, HMGCS2 overexpression significantly decreased the protein levels of non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181, indicating that all forms of Tau tested were affected by HMGCS2 in astrocytic cells. Similarly, SH-SY5Y cells transfected with HMGCS2 resulted in a prominent reduction of non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 relative to Vector (Fig. 1b). Conversely, knockdown of HMGCS2 caused a significant increase of the protein levels of non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells (Supplementary Figure 1 and Fig. 1c). These results indicated that HMGCS2 reduced Tau (non-pTau and Tau5)/pTau (pS396, pS262, pT231, pT217, and pT181) protein levels in U251MG and SH-SY5Y cells.

Fig. 1

HMGCS2 reduces the protein levels of Tau species in U251MG and SH-SY5Y cells. (a) Representative western blots (left) and bar plot summary (right) of non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector for 48 h. (b) Representative western blots (left) and bar plot summary (right) of non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in SH-SY5Y cells transfected with HMGCS2 or Vector for 48 h. (c) Representative western blots (left) and bar plot summary (right) of HMGCS2, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 siRNA or non-targeting siRNA (siControl) for 48 h. All values are normalized to control. Data are presented as means±SEM, *p < 0.05, **p < 0.01 (two-tailed independent samples Student’s t test), n = 3 in each group.

HMGCS2 did not alter the active kinases or phosphatases associated with Tau

The phosphorylation state of Tau is regulated by a balance between the activities of kinases and phosphatases [36, 37]. Tau kinases fall into two main categories: the proline-directed kinases including glycogen synthase kinase-3 Beta (GSK3β, EC 2.7.11.26) and cyclin dependent kinase 5 (CDK5, EC 2.7.11.22), and non-proline-directed kinases such as Tau-tubulin kinase (TTBK, EC 2.7.11.1), serine/threonine kinase (AKT, EC 2.7.11.1), AMP-activated protein kinase (AMPK, EC 2.7.11.1), and calcium/calmodulin dependent protein kinase II (CaMKII, EC 2.7.11.17) [38]. Among various phosphatases, protein phosphatase 2 (PP2A, EC 5.2.1.8) is the most active one and its activity is compromised in AD [39, 40]. To determine whether HMGCS2 might regulate Tau/pTau through the related enzyme activity, we assessed the protein levels of the active forms of these enzymes in cells overexpressing or silencing HMGCS2. Unexpectedly, the relative protein levels of p-AKT/AKT, p-GSK3β/GSK3β, and p-AMPK/AMPK, and those of CDK5, CAMKII, TTBK, and PP2A were not significantly changed in both U251MG and SH-SY5Y cells overexpressing HMGCS2 (Fig. 2a, b), in which the corresponding transcripts as well as the Tau itself encoding gene microtubule-associated protein Tau (MAPT) failed to show any differences relative to Vector (Supplementary Table 1). Similarly, no overt alteration of the active form of these enzymes were observed by siHMGCS2 in U251MG cells (Fig. 2c). These results indicated that regulation of the active kinases/phosphatases associated with Tau including itself MAPT transcript levels is unlikely to contribute to reduction in the Tau/pTau levels mediated by HMGCS2.

Fig. 2

HMGCS2 does not alter the protein levels of the active kinases or phosphatases associated with Tau. (a) Representative western blots (left) and bar plot summary (right) of p-AKT, AKT, p-GSK3β, GSK3β, p-AMPK, AMPK, CDK5, CaMKII, TTBK, and PP2A in U251MG cells transfected with HMGCS2 or Vector for 48 h. (b) Representative western blots (left) and bar plot summary (right) of p-AKT, AKT, p-GSK3β, GSK3β, p-AMPK, AMPK, CDK5, CaMKII, TTBK, and PP2A in SH-SY5Y cells transfected with HMGCS2 or Vector for 48 h. (c) Representative western blots (left) and bar plot summary (right) of HMGCS2, p-AKT, AKT, p-GSK3β, GSK3β, p-AMPK, AMPK, CDK5, CaMKII, TTBK, and PP2A in U251MG cells transfected with HMGCS2 siRNA or non-targeting siRNA (siControl) for 48 h. All values are normalized to control. Data are presented as means±SEM, **p < 0.01 (two-tailed independent samples Student’s t test), n = 3 in each group.

HMGCS2-mediated Tau/pTau reduction bypassed the proteasomes but was paralleled by the enhanced autophagy

It is reported that Tau isolated from AD brain is ubiquitinated, which could be considered as a signal for proteasomal degradation [41], and Tau species including Tau5 in rat brain extracts are degraded in an ubiquitin- and ATP-dependent manner in vitro [42]. To test whether the ubiquitin/proteasome system (UPS) is involved in HMGCS2-mediated Tau/pTau reduction, we measured Tau/pTau levels in U251MG cells overexpressing HMGCS2, in the presence of the proteasome inhibitor Mg132 [43]. As shown in Fig. 3a, Mg132 alone induced a negligible increase in non-pTau, Tau5, and pTau (pS396, pS262, pT231, pT217, and pT181), and was unable to prevent the HMGCS2-mediated reduction of all of the selected Tau forms. Similarly, Mg132 did not significantly increase the basal protein levels of Tau/pTau and had little to no effect on Tau/pTau reduction induced by HMGCS2 in SH-SY5Y cells (Fig. 3b). These results suggested that UPS mechanisms might play no major role in HMGCS2-mediated regulation of Tau.

Fig. 3

HMGCS2-mediated Tau reduction bypasses the proteasome but is paralleled by the enhanced autophagy. (a) Representative western blots (left) and bar plot summary (right) of non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector in the presence or absence of Mg132 for 24 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“HMGCS2” versus “Vector”); ^#p < 0.05, ^# #p < 0.01 (“HMGCS2 + MG132” versus “Vector+MG132”), one way ANOVA with LSD multiple comparison test, n = 3 in each group. (b) Representative western blots (left) and bar plot summary (right) of non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in SH-SY5Y cells transfected with HMGCS2 or Vector in the presence or absence of Mg132 for 24 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“HMGCS2” versus “Vector”); ^#p < 0.05, ^# #p < 0.01 (“HMGCS2 + MG132” versus “Vector+MG132”), one way ANOVA with LSD multiple comparison test, n = 3 in each group. (c) Representative transmission electron microscope (TEM) images (left) and bar plot summary (right) of the number of autophagic vesicles per cell in U251MG transfected with HMGCS2 or Vector for 48 h. White arrows: autophagic vesicles (AVs). N, nucleus; Vec, Vector; Hm, HMGCS2. Scale bar: 2μM (left) and 1μM (right), respectively. **p < 0.01 (two-tailed independent samples Student’s t test), n = 18 in each group. Data are presented as means±SEM.

Another mechanism in Tau degradation is autophagy-lysosomal pathway [44]. We next assessed whether HMGCS2 might influence the protein levels of microtubule-associated protein light chain 3-II (LC3-II), a widely used marker for autophagosomes [45]. In line with our previous report in HEK293 cells [24], HMGCS2 knockdown in U251MG cells and HMGCS2 overexpression in SH-SY5Y cells significantly decreased and increased the relative protein levels of LC3-II, respectively (Supplementary Figure 2). Consistently, TEM images showed that the number of autophagic vacuoles was significantly increased in U251MG cells transiently transfected with HMGCS2 compared to Vector (Fig. 3c). These results indicated that HMGCS2 promoted autophagy.

HMGCS2 promoted Tau/pTau degradation via autophagic mechanism

To further identify the potential mechanism in HMGCS2-induced autophagic clearance of Tau, we assessed the protein levels of Tau/pTau in U251MG cells overexpressing HMGCS2 and in the presence of autophagosome-lysosome fusion inhibitor CQ [46], lysosomal inhibitors BafA1, the class III PI3K inhibitor 3-MA and Mechanistic Target of Rapamycin (mTOR) inhibitor Rapa that are known to regulate autophagic flux [47]. The protein levels of Lysosomal Associated Membrane Protein 1 (LAMP1) and autophagic marker LC3-II were also measured. As shown in Fig. 4a, CQ treatment significantly increased the protein levels of LC3-II and Tau/pTau but decreased those of LAMP1, in cells transiently transfected with either Vector (lane 1 versus lane 2) or HMGCS2 (lane 3 versus lane 4), indicating that CQ inhibited autophagic degradation of Tau/pTau, regardless of HMGCS2. When comparing the protein levels of Tau/pTau in CQ treated cells (lane 2 versus lane 4), no significant changes were found, indicating that HMGCS2 lost its function in degrading Tau/pTau when autophagy machinery is inhibited. Similarly, although the protein levels of LAMP1 and LC3-II were differentially regulated by BafA1 and 3-MA (Fig. 4b, c), both of the autophagy inhibitors significantly increased the protein levels of Tau/pTau in Vector (Lane 1 versus lane 2) or in HMGCS2 (lane 3 versus lane 4); and prevented HMGCS2-induced reduction of Tau/pTau (lane 2 versus lane 4). Interestingly, the autophagic inducer Rapa showed a different effect on Tau/pTau relative to the autophagic inhibitors CQ, BafA1, and 3-MA. As shown in Fig. 4d, Rapa significantly reduced the protein levels of Tau/pTau in Vector (lane 1 versus lane 2) but not in HMGCS2 (lane 3 versus lane 4), whereas HMGCS2 overexpression-induced reduction of Tau/pTau was lost in Rapa treated cells (lane 2 versus lane 4). These results indicated that autophagic regulators prevented HMGCS2-mediated reduction of Tau/pTau proteins. In HMGCS2 overexpressing cells, the failure of Rapa to alter Tau/pTau protein levels suggested that mTOR signaling might be involved in HMGCS2-mediated regulation of Taudegradation.

Fig. 4

HMGCS2 promotes Tau degradation via autophagic mechanism. (a) Representative western blots (top) and bar plot summary (bottom) of LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector in the presence or absence of Chloroquine (CQ) for 4 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“Vector+CQ” or “HMGCS2” versus “Vector”), ^#p < 0.05, ^# #p < 0.01 (“HMGCS2 + CQ” versus “HMGCS2”), one way ANOVA with LSD multiple comparison test. (b) Representative western blots (top) and bar plot summary (bottom) of LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector in the presence or absence of Bafilomycin A1 (BafA1) for 24 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“Vector+BafA1” or “HMGCS2” versus “Vector”), ^#p < 0.05, ^# #p < 0.01 (“HMGCS2 + BafA1” versus “HMGCS2”), one way ANOVA with LSD multiple comparison test. (c) Representative western blots (top) and bar plot summary (bottom) of LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector in the presence or absence of 3-Methyladenine (3-MA) for 24 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“Vector+3-MA” or “HMGCS2” versus “Vector”), ^#p < 0.05, ^# #p < 0.01 (“HMGCS2 + 3-MA” versus “HMGCS2”), one way ANOVA with LSD multiple comparison test. (d) Representative western blots (top) and bar plot summary (bottom) of LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector in the presence or absence of Rapamycin (Rapa) for 24 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“Vector+Rapa” or “HMGCS2” versus “Vector”), one way ANOVA with LSD multiple comparison test. Data are presented as means±SEM, n = 3 in each group.

Autophagic degradation of Tau/pTau by HMGCS2 was mediated by KB

HMGCS2 functions as a control enzyme in ketogenesis [21]. Accordingly, we found that HMGCS2 overexpression increased the concentration of BHB in SH-SY5Y and U251MG cells (Supplementary Figure 3), which led us to speculate that autophagic degradation of Tau/pTau by HMGCS2 might involve KB. Thus, we assessed the protein levels of LAMP1, LC3-II, and Tau/pTau in U251MG cells supplemented with exogenous KB AcAc and BHB, in which HMGCS2 was silenced or overexpressed. As shown in Fig. 5a and 5b, the protein levels of LAMP1 and LC3-II were significantly increased, whereas those of Tau/pTau were significantly decreased in cells treated with AcAc or BHB in siControl (lane 1 versus lane 2) or in siHMGCS2 (lane 3 versus lane 4). However, in AcAc or BHB treated cells, silencing of HMGCS2 did not induce any significant changes of the protein levels of Tau/pTau (lane 2 versus lane 4). Moreover, as shown in Fig. 5c and 5d, in HMGCS2-overexpressing cells, AcAc or BHB neither increased nor reduced the protein levels of LAMP1, LC3-II, and Tau/pTau (lane 3 versus lane 4) and in AcAc or BHB treated cells, overexpression of HMGCS2 failed to alter the protein level of Tau/pTau (lane 2 versus lane 4). These results indicated that AcAc and BHB promoted autophagic degradation of Tau/pTau; and in AcAc or BHB treated cells, silencing or overexpression of HMGCS2 did not result in further alterations of Tau/pTau protein levels, suggesting that autophagic degradation of Tau/pTau by HMGCS2 was mediated by KB.

Fig. 5

Autophagic clearance of Tau by HMGCS2 is mediated by KB. (a) Representative western blots (top) and bar plot summary (bottom) of HMGCS2, LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 siRNA or non-targeting siRNA (siControl) in the presence or absence of acetoacetate (AcAc) for 24 h. Values are normalized to “siControl”. *p < 0.05, **p < 0.01 (“siControl+AcAc” or “siHMGCS2” versus “siControl”), ^#p < 0.05, ^# #p < 0.01 (“siControl+AcAc” or “siHMGCS2” versus “siHMGCS2 + AcAc”), one way ANOVA with LSD multiple comparison test. (b) Representative western blots (top) and bar plot summary (bottom) of HMGCS2, LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 siRNA or non-targeting siRNA (siControl) in the presence or absence of β-hydroxybutyrate (BHB) for 24 h. Values are normalized to “siControl”. *p < 0.05, **p < 0.01 (“siControl+BHB” or “siHMGCS2” versus “siControl”), ^#p < 0.05, ^# #p < 0.01 (“siControl+BHB” or “siHMGCS2” versus “siHMGCS2 + BHB”), one way ANOVA with LSD multiple comparison test. (c) Representative western blots (top) and bar plot summary (bottom) of LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector in the presence or absence of AcAc for 24 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“Vector+AcAc” or “HMGCS2” versus “Vector”), one way ANOVA with LSD multiple comparison test. (d) Representative western blots (top) and bar plot summary (bottom) of LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells transfected with HMGCS2 or Vector in the presence or absence of BHB for 24 h. Values are normalized to “Vector”. *p < 0.05, **p < 0.01 (“Vector+BHB” or “HMGCS2” versus “Vector”), one way ANOVA with LSD multiple comparison test. Data are presented as means±SEM, n = 3 in each group.

ANKRD42 was identified as a target gene of HMGCS2

To further understand the molecular alterations induced by HMGCS2, we performed RNA-sequencing in SH-SY5Y cells transiently overexpressing HMGCS2. A total of 43 DEGs were identified (Fig. 6a). Among these DEGs, 16 genes were upregulated, such as GREM1, TOR4A, ANKRD24, MMP1, GNLY, GOLGA6A, and RN7SKP9. 27 genes were downregulated, including FGF22, WBP1LP2, TVP23CP2, U6, and MED28P8. KEGG pathway analysis showed that the upregulated DEGs were mainly enriched in PPAR signaling pathway, whereas the downregulated DEGs were clustered in the PI3K-Akt signaling pathway (Fig. 6b). Both of the signaling pathway were correlated with autophagy [48, 49]. GO analysis revealed the molecular function of HMGCS2 in the upregulated DEGs were mainly involved in mRNA binding and structural constituent of ribosome, whereas the downregulated DEGs were associated with actin binding, transferase activity, and protein kinase activity (Fig. 6c). Further protein-protein interaction analysis identified ANKRD42 that interacted with other proteins in the network (Fig. 6d), in which ANKDR24 was associated with tuberous sclerosis 1 (TSC1) [50], an upstream negative regulator of mTOR [51]. These results indicated that HMGCS2-induced molecular alterations were associated with autophagy, in which ANKRD24 might play a role.

Fig. 6

ANKRD24 is identified as a target gene of HMGCS2. (a) Heat map of up and down-regulated DEGs regulated by HMGCS2. The red plots for positive-regulated genes and the blue plots for negative-regulated genes. (b) The KEGG pathway enrichment analysis of DEGs regulated by HMGCS2. The upregulated pathways are shown on the left, and the downregulated pathway is shown on the right. (c) The representative enriched biological processes of DEGs revealed by Go pathway analysis (left: upregulated; right: downregulated). BP, biological process; CC, cellular component; MF, molecular function; UD, the Undef. Digits denote input number. (d) The protein-protein interaction network analysis of ANKRD24.

Silencing of ANKRD24 attenuated HMGCS2-mediated degradation of Tau/pTau

Given the central role of ANKRD24 in the molecular network of HMGCS2, we argued whether ANKRD24 might directly regulate autophagy, which might be involved in HMGCS2-induced clearance of Tau. In U251MG cells, ANKRD24 siRNA was transiently transfected alone, or co-transfected with either HMGCS2 siRNA or overexpressing plasmid, and the protein levels of LAMP1/LC3-II and Tau/pTau were assessed accordingly. The optimistic ANKRD24 siRNA sequence was selected to ensure the best knockdown effect (Supplementary Figure 4). As shown in Fig. 7a, siANKRD24 significantly decreased the relative protein levels of LAMP1 and LC3-II, which was paralleled by the significant increase of those of Tau/pTau in Vector (lane1 versus lane 2) and in HMGCS2 (lane 3 versus lane 4). When comparing the effect of HMGCS2 overexpression in cells in which ANKRD24 was silenced (lane 2 versus lane 4), a significant decrease of the protein levels of Tau/pTau was found in HMGCS2. Moreover, as shown in Fig. 7b, when HMGCS2 was conversely silenced, additional ANKRD24 knockdown did not result in a significant change of the protein levels of Tau/pTau (lane 3 versus lane 4); and similarly, in cells where ANKRD24 was silenced, additional HMGCS2 knockdown failed to cause significant alterations of Tau/pTau protein levels (lane 2 versus lane 4). In line with the function of ANKRD24 in autophagy, TEM images showed that the number of autophagic vacuoles was significantly decreased in siANKRD24, whereas the increased number of autophagic vacuoles induced by HMGCS2 overexpression was diminished from above baseline (Fig. 3c) to near baseline (Fig. 7c,d) in cells co-transfected siANKRD24 with HMGCS2 plasmid. These results indicated that ANKRD24 promoted autophagic clearance of Tau/pTau, and silencing of ANKRD24 attenuated the effect of HMGCS2 on Tau degradation.

Fig. 7

ANKRD24 silencing attenuates HMGCS2-mediated degradation of Tau. (a) Representative western blots (left) and bar plot summary (right) of ANKRD24, LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells co-transfected HMGCS2 or Vector with ANKRD24 siRNA or non-targeting siRNA (siControl) for 48 h. Values are normalized to “Vector+siControl”, *p < 0.05, **p < 0.01 (“Vector+siANKRD24” or “ HMGCS2 + siControl” versus “Vector+siControl”), ^#p < 0.05, ^# #p < 0.01 (Vector+siANKRD24” or “HMGCS2 + siControl” versus “HMGCS2 + siANKRD24”), one way ANOVA with LSD multiple comparison test, n = 3 in each group. (b) Representative western blots (left) and bar plot summary (right) of HMGCS2, ANKRD24, LAMP1, LC3-II, non-pTau, Tau5, pS396, pS262, pT231, pT217, and pT181 in U251MG cells co-transfected HMGCS2 siRNA (or HMGCS2 siControl) with ANKRD24 siRNA (or ANKRD24 siControl) for 48 h. Values are normalized to “siControl+siControl”, *p < 0.05, **p < 0.01 (“siControl+siANKRD24” or “siHMGCS2 + siControl” versus “siControl+siControl”, ^#p < 0.05 (“siControl+siANKRD24” versus “siHMGCS2 + siANKRD24”), one way ANOVA with LSD multiple comparison test, n = 3 in each group. Representative transmission electron microscope (TEM) images (c) and bar blot summary (d) of the number of autophagic vesicles per cell in U251MG cotransfected ANKRD24 siRNA with HMGCS2 or Vector for 48 h. White arrows: autophagic vesicles (AVs). N, nucleus; siCtr, siControl; siANK, siANKRD24; HM, HMGCS2. Scale bar: 2μM (top) and 1μM (bottom), respectively. **p < 0.01 (“siANK” versus “siCtr”), ^# #p < 0.01 (“siANK+HM” versus “siANK”), one way ANOVA with LSD multiple comparison test, n = 18 in each group. (e) Protein levels of Aβ₄₀ and Aβ₄₂ were measured by the Enzyme-linked immunosorbent assay kits in the lysates and medium of HEK-APP cells transfected with HMGCS2 or Vector for 48 h, Values were normalized to “Vector”, *p < 0.05, **p < 0.01 (two-tailed independent samples Student’s t test), n = 4 in each group. Data are presented as means±SEM.

As HMGCS2 promotes AβPP degradation, we additionally assessed the effect of HMGCS2 on Aβ metabolism in HEK-APP cells (stably express human full-length APP695) that produce the detectable concentration of Aβ. As shown in Fig. 7e, the concentrations of extracellular and intracellular Aβ₄₀ and Aβ₄₂ were significantly reduced in cells that overexpressed HMGCS2, whereas no significant alteration of the known Aβ-degrading enzymes insulin degrading enzyme (IDE) and neprilysin was found (Supplementary Figure 5). In line with this, the corresponding transcripts was not changed in HMGCS2 overexpressing cells (Supplementary Table 1). These results indicated that HMGCS2 was an important regulator in Aβ clearance as well.

DISCUSSION

Astrocytes constitute up to 50% of the brain volume [52, 53]. Although these cells are generally regarded as housekeepers in preserving the optimal microenvironment for neuronal function and survival [54], new lines of evidence suggest that the pathological remodeling of astrocytes contributes to cognitive decline of AD [55]. Particularly, the role of astrocytic autophagy in the clearance of age-related proteins including Tau, alpha-synuclein, and Aβ has been reported [56 –58]. In our study, Tau/pTau is expressed in both astrocytes and neurons, which is in line with previous reports [59, 60]. Accordingly, we show that HMGCS2, the rate-limiting enzyme in ketogenesis, directly controls Tau/pTau degradation through autophagy in SH-SY5Y and U251MG cells, thus linking ketogenesis with the etiology of AD. Moreover, our study uncovers a novel role of ANKRD24 in autophagic degradation of Tau/pTau.

The state of pTau is tightly regulated by protein kinases and phosphatase. pTau by GSK3β occurs at 42 sites and 29 of them including pT231, pS262, pS396, pT181, and pT217 are phosphorylated in AD brains [61]. CDK5 phosphorylates Tau at 11 sites such as pS396, pT181, pT217 and pT231 [62]. AMPK is a sensor of cellular stress that maintains energy homeostasis and can directly phosphorylate Tau at pT231 and pS396 [63]. In addition to tubulin phosphorylation, TTBK dose-dependently induce Tau phosphorylation at 10 sites and Tau aggregation into NFTs [64]. Other kinase examples including CaMKII for pT217/pS262 and AKT-GSK3β for pTau have also been reported [65, 66]. Among various phosphatases, PP2A that is responsible for pT217 and pT231 predominates the phosphatase activity, which is reduced by half in the brain of AD [67]. In our study, the relative protein and transcripts levels of these kinases/phosphatases associated with Tau including itself encoding gene MAPT are not significantly altered, suggesting that HMGCS2 regulation of pTau bypasses the functional scope of phosphorylation/dephosphorylation system and directs toward the protein clearance system.

Abnormal accumulation of NFTs per se may indicate a primary defect in protein clearance [68]. The involvement of UPS in Tau degradation is supported by that in vitro brain Tau can be directly degraded by proteasome [42, 69]. Moreover, Hsc70-interacting protein (CHIP, also termed Stub1) induces ubiquitination and aggregation of Tau, whereas overexpression of Hsp70 reduces Tau protein levels [70]. Pharmacological regulation of Hsp70 controls Tau degradation in Hela cells, which can be partially blocked by proteasome inhibitor epoxomicin [71]. It is reported that proteasome inhibitor lactacystin blocks Tau degradation in SH-SY5Y cells treated with protein synthesis inhibitor cycloheximide [72]. However, subsequent studies demonstrate that inhibition of the proteasomes by lactacystin does not change Tau levels in cultured primary neurons or SH-SY5Y cells, in the presence or absence of cycloheximide for 24 h [73, 74]. Notably, the function of above-mentioned CHIP and Hsp70 might be beyond UPS, as they also regulate autophagic protein degradation [75, 76]. In our study, MG132 treatment for 24 h does not result in a significant alteration of Tau/pTau levels in both U251MG and SH-SY5Y cells. We speculate that at least in our experimental condition, UPS is less likely involved in HMGCS2-mediated degradation of Tau/pTau.

Autophagy is a highly dynamic and tightly regulated cellular event in eukaryotic cells [77]. Inhibitors of various autophagy process, such as NH₄Cl, CQ, and 3-MA, delay Tau degradation with enhanced aggregation [78]. Conversely, autophagy inducers including the mTOR inhibitor Rapa and nicotinamide promote the degradation of Tau/pTau [44, 79]. In our study, HMGCS2-induced reduction of Tau/pTau levels is abolished by the autophagic inhibitors or the inducer Rapa (lane 2 versus lane 4 in Fig. 4), supporting that autophagic machinery is employed by HMGCS2. Moreover, in cells overexpressing HMGCS2, the reduction of Tau/pTau levels is reversed by CQ, BafA1, and 3-MA but not Rapa (lane 3 versus lane 4 in Fig. 4), suggesting that mTOR signaling is involved in HMGCS2 pathway, which acts as part of autophagic flux. Interestingly, the effect of Rapa is mimicked by AcAc or BHB, which fails to further alter Tau/pTau levels in HMGCS2 overexpressing cells (lane 3 versus lane 4 in Fig. 5c and 5d), in line with the role of KB in promoting autophagy [80] by inhibition of the mTOR [24, 81]. These findings, together with that supplement of AcAc or BHB compensates for the impaired degradation of Tau/pTau by HMGCS2 silencing (Fig. 5), lead to a speculation that KB and mTOR signaling downstream of HMGCS2 mediate autophagic clearance of Tau/pTau.

ANKRD24 is expressed mainly in astrocytes and neurons of human brain [82], whereas its functional role is relatively unknown. Only two reports demonstrate that ANKRD24 acts as a structural component of the stereocilia rootlet of sensory hair cells, and the RNA levels are altered in neuroblastoma [83, 84]. In our study, we identify ANKRD24 as a target gene of HMGCS2, which is verified by western blotting analysis in U251MG cells overexpressing or silencing HMGCS2 (Fig. 7a, b). Knockdown of ANKRD24 decreases the relative protein levels of LAPM1/LC3II but increases those of Tau/pTau, with the reduced formation of autophagic vacuoles, indicating that ANKRD24 regulates autophagy. Moreover, whereas HMGCS2-induced reduction of Tau/pTau levels is attenuated by ANKRD24 silencing, the augmentation of Tau/pTau levels by HMGCS2 knockdown cannot be further affected by ANKRD24 silencing (Fig. 7b), suggesting that ANKRD24 is involved in the autophagic pathway of HMGCS2. Although the potential mechanisms are currently unclear, protein-protein interaction analysis reveals that ANKRD24 interacts with TSC1, another autophagy-related protein that could inhibit mTOR [85]. Importantly, TSC1 is functionally linked to Tau accumulation in neurons, and loss of TSC1 averts Tau clearance through chaperon-mediated autophagy [86, 87]. Given the role of TSC1 in mTOR signaling [88], and Rapa prevents HMGCS2-induced regulation of Tau/pTau (Fig. 4), we speculate that the function of ANKRD24 in autophagy might be associated with mTOR, which remains to be studied in the future.

In summary, our data demonstrate that HMGCS2 significantly decreases Tau/pTau levels through autophagic pathway in both astrocytic and neuronal cell lines. We provide evidence that KB and ANKRD24 regulate autophagy and are involved in HMGCS2-mediated degradation of Tau/pTau. We propose that KB downstream of HMGCS2 activates autophagic pathway through mTOR, whereas ANKRD24 might work together with mTOR and leads to the enhanced degradation of Tau/pTau (Fig. 8). Together with our previous study that HMGCS2 promotes autophagic clearance of AβPP, HMGCS2 is a multifunctional molecule that inhibits both Tau deposition and amyloidogenesis, supporting HMGCS2 as a potential target for AD treatment.

Fig. 8

Schematic diagram depicts the potential mechanism underlying HMGCS2-mediated degradation of Tau/pTau. HMGCS2 enhances KB generation and promotes autophagic degradation of Tau/pTau in an mTOR dependent manner, whereas the increased expression of ANKRD24 by HMGCS2 participates in autophagic regulation possibly through mTOR signaling.

Footnotes

ACKNOWLEDGMENTS

This work was supported by NSFC (81971030) and Chongqing Education commission (KJZD-K201900404) to G-J Chen and by The Strategic Science & Technology Cooperation Project of North Sichuan Medical College and Nanchong Municipal Government (18SXHZ0184), Scientific Research Development Project of North Sichuan Medical College (CBY17-A-ZD03) and Sichuan Provincial Health and Family Planning Commission (18PJ335) to Li-Tian Hu.

Authors’ disclosures available online ().

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

References

Fiandaca

, Mapstone

, Cheema

, Federoff

(2014) The critical need for defining preclinical biomarkers in Alzheimer’s disease. Alzheimers Dement 10, S196–212.

Jakob-Roetne

, Jacobsen

(2009) Alzheimer’s disease: From pathology to therapeutic approaches. Angew Chem Int Ed Engl 48, 3030–3059.

Scheltens

, De Strooper

, Kivipelto

, Holstege

, Chételat

, Teunissen

, Cummings

, van der Flier

(2021) Alzheimer’s disease. Lancet 397, 1577–1590.

Karran

, De Strooper

(2022) The amyloid hypothesis in Alzheimer disease: New insights from new therapeutics. Nat Rev Drug Discov 21, 306–318.

Arnsten

AFT

, Datta

, Del Tredici

, Braak

(2021) Hypothesis: Tau pathology is an initiating factor in sporadic Alzheimer’s disease. Alzheimers Dement 17, 115–124.

Bloom

(2014) Amyloid-β and Tau: The trigger and bullet in Alzheimer disease pathogenesis. JAMA Neurol 71, 505–508.

Alonso

, Cohen

, Corbo

, Morozova

, ElIdrissi

, Phillips

, Kleiman

(2018) Hyperphosphorylation of tau associates with changes in its function beyond microtubule stability. Front Cell Neurosci 12, 338.

Eftekharzadeh

, Daigle

, Kapinos

, Coyne

, Schiantarelli

, Carlomagno

, Cook

, Miller

, Dujardin

, Amaral

, Grima

, Bennett

, Tepper

, DeTure

, Vanderburg

, Corjuc

, DeVos

, Gonzalez

, Chew

, Vidensky

, Gage

, Mertens

, Troncoso

, Mandelkow

, Salvatella

, Lim

RYH

, Petrucelli

, Wegmann

, Rothstein

, Hyman

(2018) Tau protein disrupts nucleocytoplasmic transport in Alzheimer’s disease, Neuron 99, 925–940.e927.

Selkoe

, Hardy

(2016) The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol Med 8, 595–608.

10.

Alonso Adel

, Mederlyova

, Novak

, Grundke-Iqbal

, Iqbal

(2004) Promotion of hyperphosphorylation by frontotemporal dementia Tau mutations. J Biol Chem 279, 34873–34881.

11.

, Zhang

, Li

, Run

, Liang

, Li

, Liu

, Lee

, Grundke-Iqbal

, Iqbal

, Vocadlo

, Liu

, Gong

(2012) Differential effects of an O-GlcNAcase inhibitor on Tau phosphorylation, PLoS One 7, e35277.

12.

Wennstrom

, Janelidze

, Nilsson

KPR

, Netherlands Brain

, Serrano

, Beach

, Dage

, Hansson

(2022) Cellular localization of p-Tau217 in brain and its association with p-Tau217 plasma levels. Acta Neuropathol Commun 10, 3.

13.

Thijssen

, La Joie

, Strom

, Fonseca

, Iaccarino

, Wolf

, Spina

, Allen

, Cobigo

, Heuer

, VandeVrede

, Proctor

, Lago

, Baker

, Sivasankaran

, Kieloch

, Kinhikar

, Yu

, Valentin

, Jeromin

, Zetterberg

, Hansson

, Mattsson-Carlgren

, Graham

, Blennow

, Kramer

, Grinberg

, Seeley

, Rosen

, Boeve

, Miller

, Teunissen

, Rabinovici

, Rojas

, Dage

, Boxer

(2021) Plasma phosphorylated Tau 217 and phosphorylated Tau 181 asbiomarkers in Alzheimer’s disease and frontotemporal lobardegeneration: A retrospective diagnostic performance study. Lancet Neurol 20, 739–752.

14.

Johnson

, Stoothoff

(2004) Tau phosphorylation in neuronal cell function and dysfunction. J Cell Sci 117, 5721–5729.

15.

Jin

, Shepardson

, Yang

, Chen

, Walsh

, Selkoe

(2011) Soluble amyloid beta-protein dimers isolated from Alzheimer cortex directly induce Tau hyperphosphorylation and neuritic degeneration. Proc Natl Acad Sci U S A 108, 5819–5824.

16.

Gotz

, Xia

, Leinenga

, Chew

, Nicholas

(2013) What renders tau toxic. Front Neurol 4, 72.

17.

Šimić

, Babić Leko

, Wray

, Harrington

, Delalle

, Jovanov-Milošević

, Bažadona

, Buée

, de Silva

, Di Giovanni

, Wischik

, Hof

(2016) Tau protein hyperphosphorylation and aggregation in Alzheimer’s disease and other tauopathies, and possible neuroprotective strategies. Biomolecules 6, 6.

18.

Jangampalli Adi

, Reddy

(2021) Phosphorylated Tau targeted small-molecule PROTACs for the treatment of Alzheimer’s disease and Tauopathies. Biochim Biophys Acta Mol Basis Dis 1867, 166162.

19.

Wang

, Mandelkow

(2016) Tau in physiology and pathology. Nat Rev Neurosci 17, 5–21.

20.

Meier

, Bell

, Lyons

, Rodriguez-Rivera

, Ingram

, Fontaine

, Mechas

, Chen

, Wolozin

, LeVine

3rd , Zhu

, Abisambra

(2016) Pathological tau promotes neuronal damage by impairing ribosomal function and decreasing protein synthesis. J Neurosci 36, 1001–1007.

21.

Hegardt

(1999) Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase: A control enzyme in ketogenesis. Biochem J 338 (Pt 3), 569–582.

22.

Jebb

, Hiller

(2018) Recurrent loss of HMGCS2 shows that ketogenesis is not essential for the evolution of large mammalian brains, Elife 7, e38906.

23.

Yang

, Long

, Xu

, Zhu

, Deng

, Yan

, Sun

, Chen

(2019) Age- and nicotine-associated gene expression changes in the hippocampus of APP/PS1 mice. J Mol Neurosci 69, 608–622.

24.

, Zhu

, Lai

, Long

, Zha

, Hu

, Zhang

, Chen

(2017) HMGCS2 promotes autophagic degradation of the amyloid-beta precursor protein through ketone body-mediated mechanisms. Biochem Biophys Res Commun 486, 492–498.

25.

Newport

, VanItallie

, Kashiwaya

, King

, Veech

(2015) A new way to produce hyperketonemia: Use of ketone ester in a case of Alzheimer’s disease. Alzheimers Dement 11, 99–103.

26.

Kashiwaya

, Bergman

, Lee

, Wan

, King

, Mughal

, Okun

, Clarke

, Mattson

, Veech

(2013) A ketone ester diet exhibits anxiolytic and cognition-sparing properties, and lessens amyloid and Tau pathologies in a mouse model of Alzheimer’s disease. Neurobiol Aging 34, 1530–1539.

27.

Lilamand

, Mouton-Liger

, Paquet

(2021) Ketogenic diet therapy in Alzheimer’s disease: An updated review. Curr Opin Clin Nutr Metab Care 24, 372–378.

28.

Alvarez-Erviti

, Seow

, Schapira

, Gardiner

, Sargent

, Wood

, Cooper

(2011) Lysosomal dysfunction increases exosome-mediated alpha-synuclein release and transmission. Neurobiol Dis 42, 360–367.

29.

Ertmer

, Huber

, Gilch

, Yoshimori

, Erfle

, Duyster

, Elsasser

, Schatzl

(2007) The anticancer drug imatinib induces cellular autophagy. Leukemia 21, 936–942.

30.

Zhu

, Long

, Luo

, Yan

, Lai

, Zhao

, Zhou

, Wang

, Shen

, Liu

, Deng

, Wang

, Sun

, Chen

(2019) MMP13 inhibition rescues cognitive decline in Alzheimer transgenic mice via BACE1 regulation. Brain 142, 176–192.

31.

Noh

, Hah

, Nilufar

, Han

, Bong

, Kang

, Cho

, Choi

(2006) Acetoacetate protects neuronal cells from oxidative glutamate toxicity. J Neurosci Res 83, 702–709.

32.

Bonvento

, Bolaños

(2021) Astrocyte-neuron metabolic cooperation shapes brain activity. Cell Metab 33, 1546–1564.

33.

Kovacs

, Ferrer

, Grinberg

, Alafuzoff

, Attems

, Budka

, Cairns

, Crary

, Duyckaerts

, Ghetti

, Halliday

, Ironside

, Love

, Mackenzie

, Munoz

, Murray

, Nelson

, Takahashi

, Trojanowski

, Ansorge

, Arzberger

, Baborie

, Beach

, Bieniek

, Bigio

, Bodi

, Dugger

, Feany

, Gelpi

, Gentleman

, Giaccone

, Hatanpaa

, Heale

, Hof

, Hofer

, Hortobagyi

, Jellinger

, Jicha

, Ince

, Kofler

, Kovari

, Kril

, Mann

, Matej

, McKee

, McLean

, Milenkovic

, Montine

, Murayama

, Lee

, Rahimi

, Rodriguez

, Rozemuller

, Schneider

, Schultz

, Seeley

, Seilhean

, Smith

, Tagliavini

, Takao

, Thal

, Toledo

, Tolnay

, Troncoso

, Vinters

, Weis

, Wharton

, White

3rd , Wisniewski

, Woulfe

, Yamada

, Dickson

(2016) Aging-related Tau astrogliopathy (ARTAG): Harmonized evaluation strategy. Acta Neuropathol 131, 87–102.

34.

Brickman

, Manly

, Honig

, Sanchez

, Reyes-Dumeyer

, Lantigua

, Lao

, Stern

, Vonsattel

, Teich

, Airey

, Proctor

, Dage

, Mayeux

(2021) Plasma p-Tau181, p-Tau217, and other blood-based Alzheimer’s disease biomarkers in a multi-ethnic, community study. Alzheimers Dement 17, 1353–1364.

35.

Xiang

, Song

, Deng

, Tang

, Min

, Luo

, Wen

, Li

, Chen

, Ma

, Zhu

, Yan

, Chen

(2019) Mitochondrial methionine sulfoxide reductase B2 links oxidative stress to Alzheimer’s disease-like pathology. Exp Neurol 318, 145–156.

36.

Martin

, Latypova

, Wilson

, Magnaudeix

, Perrin

, Yardin

, Terro

(2013) Tau protein kinases: Involvement in Alzheimer’s disease. Ageing Res Rev 12, 289–309.

37.

Tapia-Rojas

, Cabezas-Opazo

, Deaton

, Vergara

, Johnson

GVW

, Quintanilla

(2019) It’s all about Tau. Prog Neurobiol 175, 54–76.

38.

Oliveira

, Costa

, de Almeida

MSC

, da Cruz

ESOAB

, Henriques

(2017) Protein phosphorylation is a key mechanism in Alzheimer’s disease. J Alzheimers Dis 58, 953–978.

39.

Gong

, Shaikh

, Wang

, Zaidi

, Grundke-Iqbal

, Iqbal

(1995) Phosphatase activity toward abnormally phosphorylated Tau: Decrease in Alzheimer disease brain. J Neurochem 65, 732–738.

40.

Taleski

, Sontag

(2018) Protein phosphatase 2A and Tau: An orchestrated ‘Pas de Deux’. FEBS Lett 592, 1079–1095.

41.

Pickart

(2001) Ubiquitin enters the new millennium. Mol Cell 8, 499–504.

42.

Zhang

, Liu

, Li

, Wang

(2005) Microtubule-associated protein Tau is a substrate of ATP/Mg(2+)-dependent proteasome protease system. J Neural Transm (Vienna) 112, 547–555.

43.

Kisselev

(2021) Site-specific proteasome inhibitors. Biomolecules 12, 54.

44.

Hamano

, Enomoto

, Shirafuji

, Ikawa

, Yamamura

, Yen

, Nakamoto

(2021) Autophagy and tau protein. Int J Mol Sci 22, 7475.

45.

Mizushima

, Yoshimori

, Levine

(2010) Methods in mammalian autophagy research. Cell 140, 313–326.

46.

Mauthe

, Orhon

, Rocchi

, Zhou

, Luhr

, Hijlkema

, Coppes

, Engedal

, Mari

, Reggiori

(2018) Chloroquine inhibits autophagic flux by decreasing autophagosome-lysosome fusion. Autophagy 14, 1435–1455.

47.

Kocaturk

, Akkoc

, Kig

, Bayraktar

, Gozuacik

, Kutlu

(2019) Autophagy as a molecular target for cancer treatment. Eur J Pharm Sci 134, 116–137.

48.

Gou

, Jiang

, Zhang

, Xu

, Zhang

, Shi

, Hou

(2020) PPARdelta is a regulator of autophagy by its phosphorylation. Oncogene 39, 4844–4853.

49.

Heras-Sandoval

, Perez-Rojas

, Hernandez-Damian

, Pedraza-Chaverri

(2014) The role of PI3K/AKT/mTOR pathway in the modulation of autophagy and the clearance of protein aggregates in neurodegeneration. Cell Signal 26, 2694–2701.

50.

Sakai

, Shaw

, Dawson

, Dugas

, Al-Mohtaseb

, Hill

, Zoghbi

(2011) Protein interactome reveals converging molecular pathways among autism disorders, Sci Transl Med 3, 86ra49.

51.

Gao

, Zhang

, Arrazola

, Hino

, Kobayashi

, Yeung

, Ru

, Pan

(2002) Tsc tumour suppressor proteins antagonize amino-acid-TOR signalling. Nat Cell Biol 4, 699–704.

52.

Brambilla

, Martorana

, Rossi

(2013) Astrocyte signaling and neurodegeneration: New insights into CNS disorders. Prion 7, 28–36.

53.

Tower

, Young

(1973) The activities of butyrylcholinesterase and carbonic anhydrase, the rate of anaerobic glycolysis, and the question of a constant density of glial cells in cerebral cortices of various mammalian species from mouse to whale. J Neurochem 20, 269–278.

54.

Sofroniew

, Vinters

(2010) Astrocytes: Biology and pathology. Acta Neuropathol 119, 7–35.

55.

Pekny

, Pekna

, Messing

, Steinhäuser

, Lee

, Parpura

, Hol

, Sofroniew

, Verkhratsky

(2016) Astrocytes: A central element in neurological diseases. Acta Neuropathol 131, 323–345.

56.

Sung

, Jimenez-Sanchez

(2020) Autophagy in Astrocytes and its Implications in Neurodegeneration. J Mol Biol 432, 2605–2621.

57.

Wang

, Xu

(2020) Astrocytes autophagy in aging and neurodegenerative disorders. Biomed Pharmacother 122, 109691.

58.

Pras

, Nollen

EAA

(2021) Regulation of age-related protein toxicity. Front Cell Dev Biol 9, 637084.

59.

Richetin

, Steullet

, Pachoud

, Perbet

, Parietti

, Maheswaran

, Eddarkaoui

, Begard

, Pythoud

, Rey

, Caillierez

, K

, Halliez

, Bezzi

, Buee

, Leuba

, Colin

, Toni

, Deglon

(2020) Tau accumulation in astrocytes of the dentate gyrus induces neuronal dysfunction and memory deficits in Alzheimer’s disease. Nat Neurosci 23, 1567–1579.

60.

Mate de Gerando

, d’Orange

, Augustin

, Josephine

, Auregan

, Gaudin-Guerif

, Guillermier

, Herard

, Stimmer

, Petit

, Gipchtein

, Jan

, Escartin

, Selingue

, Carvalho

, Blum

, Brouillet

, Hantraye

, Gaillard

, Bonvento

, Bemelmans

, Cambon

(2021) Neuronal Tau species transfer to astrocytes and induce their loss according to Tau aggregation state. Brain 144, 1167–1182.

61.

Hanger

, Anderton

, Noble

(2009) Tau phosphorylation: The therapeutic challenge for neurodegenerative disease. Trends Mol Med 15, 112–119.

62.

Patrick

, Zukerberg

, Nikolic

, de la Monte

, Dikkes

, Tsai

(1999) Conversion of p35 to p25 deregulates Cdk5 activity and promotes neurodegeneration. Nature 402, 615–622.

63.

Vingtdeux

, Davies

, Dickson

, Marambaud

(2011) AMPK is abnormally activated in tangle- and pre-tangle-bearing neurons in Alzheimer’s disease and other Tauopathies. Acta Neuropathol 121, 337–349.

64.

, Sato

, Okuyama

, Swan

, Jacobsen

, Strunk

, Ikezu

(2010) Tau-tubulin kinase 1 enhances prefibrillar Tau aggregation and motor neuron degeneration in P301L FTDP-17 Tau-mutant mice. FASEB J 24, 2904–2915.

65.

Wang

, Xia

, Grundke-Iqbal

, Iqbal

(2013) Abnormal hyperphosphorylation of Tau: Sites, regulation, and molecular mechanism of neurofibrillary degeneration, J Alzheimers Dis 33 (Suppl 1), S123–139.

66.

Chu

, Lauretti

, Praticò

(2017) Caspase-3-dependent cleavage of Akt modulates Tau phosphorylation via GSK3β kinase: Implications for Alzheimer’s disease. Mol Psychiatry 22, 1002–1008.

67.

Martin

, Latypova

, Wilson

, Magnaudeix

, Perrin

, Terro

(2013) Tau protein phosphatases in Alzheimer’s disease: The leading role of PP2A. Ageing Res Rev 12, 39–49.

68.

Chung

, Hernandez

, Sproul

, Yu

(2019) Alzheimer’s disease and the autophagic-lysosomal system. Neurosci Lett 697, 49–58.

69.

Cardozo

, Michaud

(2002) Proteasome-mediated degradation of Tau proteins occurs independently of the chymotrypsin-like activity by a nonprocessive pathway. Arch Biochem Biophys 408, 103–110.

70.

Petrucelli

, Dickson

, Kehoe

, Taylor

, Snyder

, Grover

, De Lucia

, McGowan

, Lewis

, Prihar

, Kim

, Dillmann

, Browne

, Hall

, Voellmy

, Tsuboi

, Dawson

, Wolozin

, Hardy

, Hutton

(2004) CHIP and Hsp70 regulate Tau ubiquitination, degradation and aggregation. Hum Mol Genet 13, 703–714.

71.

Jinwal

, Miyata

, Koren

3rd , Jones

, Trotter

, Chang

, O’Leary

, Morgan

, Lee

, Shults

, Rousaki

, Weeber

, Zuiderweg

, Gestwicki

, Dickey

(2009) Chemical manipulation of hsp70 ATPase activity regulates Tau stability. J Neurosci 29, 12079–12088.

72.

David

, Layfield

, Serpell

, Narain

, Goedert

, Spillantini

(2002) Proteasomal degradation of Tau protein. J Neurochem 83, 176–185.

73.

Brown

, Bondada

, Keller

, Thorpe

, Geddes

(2005) Proteasome or calpain inhibition does not alter cellular Tau levels in neuroblastoma cells or primary neurons. J Alzheimers Dis 7, 15–24.

74.

Wang

, Martinez-Vicente

, Kruger

, Kaushik

, Wong

, Mandelkow

, Cuervo

, Mandelkow

(2009) Tau fragmentation, aggregation and clearance: The dual role of lysosomal processing. Hum Mol Genet 18, 4153–4170.

75.

Sha

, Rao

, Settembre

, Ballabio

, Eissa

(2017) STUB1 regulates TFEB-induced autophagy-lysosome pathway. EMBO J 36, 2544–2552.

76.

Fernández-Fernández

, Gragera

, Ochoa-Ibarrola

, Quintana-Gallardo

, Valpuesta

(2017) Hsp70 - a master regulator in protein degradation. FEBS Lett 591, 2648–2660.

77.

Ohsumi

(2014) Historical landmarks of autophagy research. Cell Res 24, 9–23.

78.

Wang

, Martinez-Vicente

, Kruger

, Kaushik

, Wong

, Mandelkow

, Cuervo

, Mandelkow

(2010) Synergy and antagonism of macroautophagy and chaperone-mediated autophagy in a cell model of pathological Tau aggregation. Autophagy 6, 182–183.

79.

Lee

, Lee

, Rubinsztein

(2013) Tau degradation: The ubiquitin-proteasome system versus the autophagy-lysosome system. Prog Neurobiol 105, 49–59.

80.

Camberos-Luna

, Gerónimo-Olvera

, Montiel

, Rincon-Heredia

, Massieu

(2016) The ketone body, β-hydroxybutyrate stimulates the autophagic flux and prevents neuronal death induced by glucose deprivation in cortical cultured neurons. Neurochem Res 41, 600–609.

81.

McDaniel

, Rensing

, Thio

, Yamada

, Wong

(2011) The ketogenic diet inhibits the mammalian target of rapamycin (mTOR) pathway. Epilepsia 52, e7–11.

82.

Darmanis

, Sloan

, Zhang

, Enge

, Caneda

, Shuer

, Hayden

Gephart MG

, Barres

, Quake

(2015) A survey of human brain transcriptome diversity at the single cell level. Proc Natl Acad Sci U S A 112, 7285–7290.

83.

Krey

, Liu

, Belyantseva

, Bateschell

, Dumont

, Goldsmith

, Chatterjee

, Morrill

, Fedorov

, Foster

, Kim

, Nuttall

, Jones

, Choi

, Friedman

, Ricci

, Zhao

, Barr-Gillespie

(2022) ANKRD24 organizes TRIOBP to reinforce stereocilia insertion points. J Cell Biol 221, e202109134.

84.

Zhang

, Ma

, Ke

, Liu

, Li

, Liu

, Wang

(2021) Development and validation of a five-RNA-based signature and identification of candidate drugs for neuroblastoma. Front Genet 12, 685646.

85.

Henske

, Jóźwiak

, Kingswood

, Sampson

, Thiele

(2016) Tuberous sclerosis complex. Nat Rev Dis Primers 2, 16035.

86.

Alquezar

, Schoch

, Geier

, Ramos

, Scrivo

, Li

, Argouarch

, Mlynarski

, Dombroski

, DeTure

, Dickson

, Yokoyama

, Cuervo

, Burlingame

, Schellenberg

, Miller

, Kao

(2021) TSC1 loss increases risk for tauopathy by inducing tau acetylation and preventing tau clearance via chaperone-mediated autophagy, Sci Adv 7, eabg3897.

87.

Sanchez

, Acker

, Gray

, Varadarajan

, Song

, Cochran

, Paula

, Lindeman

, An

, McAllister

, Alford

, Reece-Hoyes

, Russ

, Craig

, Capre

, Doherty

, Hoffman

, Luchansky

, Polydoro

, Dolmetsch

, Elwood

(2021) Genome-wide CRISPR screen identifies protein pathways modulating Tau protein levels in neurons. Commun Biol 4, 736.

88.

Lam

, Siroky

, Henske

(2018) Renal disease in tuberous sclerosis complex: Pathogenesis and therapy. Nat Rev Nephrol 14, 704–716.

HMGCS2-Induced Autophagic Degradation of Tau Involves Ketone Body and ANKRD24

Abstract

Background:

Objective:

Methods:

Results:

Conclusion:

Keywords

INTRODUCTION

MATERIALS AND METHODS

Antibodies and reagents

Cell culture and pharmacologic treatments

Plasmid and small interfering RNA (siRNA) transfection

Western blotting analysis

RNA sequencing and bioinformation analysis

Transmission electron microscopy (TEM)

Enzyme-linked immunosorbent assay for Aβ40 and Aβ42

Statistical analysis

RESULTS

HMGCS2 reduced the protein levels of different forms of Tau protein

HMGCS2 did not alter the active kinases or phosphatases associated with Tau

HMGCS2-mediated Tau/pTau reduction bypassed the proteasomes but was paralleled by the enhanced autophagy

HMGCS2 promoted Tau/pTau degradation via autophagic mechanism

Autophagic degradation of Tau/pTau by HMGCS2 was mediated by KB

ANKRD42 was identified as a target gene of HMGCS2

Silencing of ANKRD24 attenuated HMGCS2-mediated degradation of Tau/pTau

DISCUSSION

Footnotes

ACKNOWLEDGMENTS

References

Enzyme-linked immunosorbent assay for Aβ₄₀ and Aβ₄₂