Chemicals and antibodies

MG132 (S2619), Chloroquine (CQ, DC661), and Cycloheximide (CHX, S7418) were from Selleckchem (Houston, TX, USA). Rapamycin was from LC Laboratories (Woburn, MA, USA). DMSO (196,055) was from Sigma (St. Louis, MO, USA). Corn oil was from Aladdin (Shanghai, CA). J3 was synthesized as described in Fig. 1P.

Antibodies used in this work were as follows. HTT (MAB2166) and DARPP-32 (AB10518) were from Millipore (Temecula, CA, USA). GFP (sc-9996) and ubiquitin (sc-8017) were from Santa Cruz (Dallas, TX, USA). LC3B (L7543) and β-actin were from Sigma (St. Louis, MO, USA). GAPDH (60,004–1) was from Proteintech (Wuhan, China). ATG4B (M134), ATG5 (M153), ATG16L1 (M150), LC3 (PM036), and p62 (PM045) were from MBL (Woburn, MA, USA). MTOR (2983), p-mTOR (5536), S6K (2708), p-S6K (9208, 9234), ULK1 (4773), and p-ULK1 (5869, 6888) were from Cell Signaling Technology (Danvers, MA, USA). Secondary antibodies conjugated with Alexa Fluor 488 or DyLight 594 secondary antibodies [710, 369, 35, 560] were from ThermoFisher Scientific (Carlsbad, CA, USA).

Cell culture and plasmid transfection

ATG5-deficient MEFs, ATG4B-deficient HeLa cells have been described previously [39, 40], ATG16L1-deficient HeLa cells were constructed by Cripsr/Cas9 (sgRNA primers: 5’TCGCGGTGGTTACGCTCGCT3’ #1, 5’ CAGTGTGAGCGGCGCCGGTG 3’ #2). All cells were cultured in DMEM (SH3024301) (Thermo Scientific, Rockford, IL, USA) supplied with 10% (v/v) fetal bovine serum (FBS, BI, 1752054) and 0.5% Penicillin/Streptomycin (Gibco, Life technologies, 15140163) in 5% (v/v) CO2 incubator at 37℃. HeLa cell stably expressing exon1HTT-103QmCFP (CFP-103Q) was a kind gift from professor Ai Yamamoto (Columbia University) [24]. Cells were cultured in 6-well plates to about 80% density before transfection, then 2 μg of plasmid was transfected to cells by transfection reagent Lipo2000 (Invitrogen, Carlsbad, CA, USA).

Protein extraction and immunoblot assay

For mouse brain tissue, the mouse striatum was dissected on ice and cut into pieces and sonicated for 20 times at 20% power, and lysed on ice for 1 h in brain lysis buffer (50 mM Tris, 250 mM NaCl, 5 mM EDTA, 1% Triton X-100 pH7.4) plus protease inhibitor cocktail. Then the samples were centrifuged at 16000×g at 4 ℃ for 20 min. The supernatant of the striatum sample was collected for further analysis.

The fractionation for resolution of soluble and insoluble Huntingtin in the HD cell model was carried out as previous described [41]. Briefly, cells were lysed on ice and then centrifuged at 15000×g for 20 min at 4 °C. The supernatant was collected as the soluble part. The pellet was washed three times by the lysis buffer for 5 min, and the lysis buffer supplemented with 4% SDS was added followed by 30 s sonication which was collected as the insoluble part. The protein concentration of both parts was determined by BCA Protein Quantitation Kit. 30 µg of the soluble part was resolved on 8% SDS-PAGE gels, and 10 µg of the insoluble part was resolved on 3% Tris–acetate polyacrylamide gels for very high molecular mass [42]. Alternatively, another method was applied to simplify the detection of mHTT. In this simplified method, cells were lysed on ice and then centrifuged at 15000×g for 20 min at 4 °C. The supernatant was collected as the soluble part. The pellet was washed by buffer containing 2% SDS followed by sonication. After sonication and centrifugation, the supernatant was also transferred to the soluble fraction. The pellet was thought as the insoluble protein, then 5× Loading Buffer was added to both samples and boiled for 5 min. The protein concentration of the soluble part was determined by BCA Protein Quantitation and the insoluble fractions were collected without BCA quantification, but by controlling the number of cells in each group. The soluble and insoluble fractions were loaded separately into 10% SDS-PAGE. These two methods to measure the mHTT were both applied in this study.

For immunoblot, cells were lysed by lysis buffer RIPA (Beyond time, CA) supplied with protease inhibitor cocktail (Bimake, B14012) for 10 min, then the cell lysate was centrifuged for 15 min at 12,000 rpm, 4 ℃, the supernatant was collected and quantified using BCA Protein Quantitation Kit (Thermo, WA319620). Then 30 μg of cell lysate was loaded on 8% or 12% SDS-PAGE gels and transferred to PVDF membranes (Millipore, ISEQ00010). After blocking by 5% nonfat milk and incubation with primary and secondary antibodies, specific proteins were detected using High-sig ECL Western Blotting Substrate (Tanon, 180–5001). Images were captured by Tannon Image Station 4000.

Immunostaining analysis

For cell samples, cells were cultured at 24-well confocal plates, after washing by PBS for two times, and fixed by 4% paraformaldehyde (PFA, pH = 7.4) for 15 min at room temperature, then permeabilized by 0.1% Triton X-100 for 15 min. The samples were blocked with goat serum for 30 min. Primary antibodies were added and incubated at 4 °C overnight, then followed by secondary antibodies for another 1 h at room temperature.

For mouse brain slices, mice were anesthetized and sacrificed, then perfused with 4% PFA for 48 h at 4 °C. Then the tissues were incubated in 15% sucrose for 24 h and then in 30% sucrose for 48 h at 4 °C and sectioned into 30 μm thick slices. The subsequent experiments were performed similarly to the cell experiment mentioned above except using 5% BSA for blocking. Primary antibodies HTT (1:200), DARPP-32 (1:5000), Ub (1:200) were applied followed by incubation with secondary antibodies. Images of above samples were captured by EVOS FL Auto (Life Technologies, Bothell, WA, USA) and analyzed blindly by Image J for aggregates quantifications. Fluorescence images were also taken by EVOS FL Auto or Olympus confocal microscopes (FV3000). For manual quantification of the puncta formation, at least 3 optical fields with more than 50 cells per experimental condition were analyzed. GFP-LC3 dots of cells were counted from different groups. Data from repeated experiments were subjected to statistical analysis.

Long-lived protein degradation

Long-lived protein degradation assay was carried out as described previously [43], Briefly, MEFs were cultured in 24-well plates overnight, L-[14C]-valine (PerkinElmer, NEC291EU050UC) was added to a final concentration of 0.2 µCi/ml to label intracellular proteins. Cells were incubated for 18–24 h before changing to fresh medium for another hour with 10% cold l-valine to deplete labeled short-lived proteins. The cells were then incubated in EBSS or DMEM (plus 0.1% of BSA and 10 mM of valine) with or without J3 for an additional 6 or 16 h. The culture medium was recovered, from which the degraded long-lived proteins were measured via liquid scintillation.

J3 synthesis

Benzoyl isothiocyanate (0.82 g, 5.0 mmol) and diethyl 5-amino-3-methylthiophene-2,4-dicarboxylate (1.54 g, 6.0 mmol) were dissolved in anhydrous MeCN (30 mL) and stirred at reflux. After completion of the reaction, monitored by thin layer chromatography (TLC), it was then cooled to room temperature and evaporated under vacuum. The residue was purified by column chromatography to afford diethyl 5-(3-benzoylthioureido)-3-methy-lthiophene-2,4-dicarboxylate (1.47 g, mmol, 70.0% yield) as a yellow powder. 1H NMR (400 MHz, CDCl3) δ 14.93 (s, 1H), 9.23 (s, 1H), 8.07 – 7.89 (m, 2H), 7.67 (t, J = 7.4 Hz, 1H), 7.56 (t, J = 7.7 Hz, 2H), 4.55 (q, J = 7.1 Hz, 2H), 4.37 (q, J = 7.1 Hz, 2H), 2.83 (s, 3H), 1.44 (dt, J = 19.4, 7.1 Hz, 6H). ESI–MS m/z: 421.2 [M + H]+.

Mouse model and genotyping

Huntington’s disease knock-in mouse model expression 140Q (HdhQ140) knock-in mice were kindly gifted from Professor Marian Difiglia (MassGeneral Hospital). The generation and characterization of Hdh140Q knock-in mice have been described previously [38]. Specifically, the mice’s genetic background was derived from C57BL/6, and exon1 of mice HTT was replaced by human HTT that containing 140 CAG repeats. The homozygous of HdhQ140 knock-in mice were mating with wild-type mice in male: female ration of 1:2. The first generation was heterozygotes and mating between littermates male and female mice to generate homozygous and wild-type mice. After genotyping, mice were kept for corresponding months. Homozygous mice were selected after 4 generations of Huntington’s disease mice cross the same genetic background. Mice were group-housed by 5 adults per cage with 12 h light/dark cycle. The mouse experiments were carried out following the general guidelines published by the Association for Assessment and Accreditation of Laboratory Animal Care. The Animal Care and Use Committee of the School of Medicine at Guangzhou University of Chinese medicine approved the protocol used in animal experiments (No.20190704006). Mouse brain tissue was acutely dissected for protein extraction and immunohistology analysis.

When the mice grew to 6 weeks, 2–3 mm of mouse tail was cropped and collected in tubes. After extraction of DNA, PCR reaction buffer was prepared with primers (F: 5′ CTGCACCGACCGTGAGTCC 3′, R: 5′ GAAGGCACTGGAGTCGTGAC 3′), then the amplification system was operated and separated by 2% agarose. Under gel imaging system, two bands appeared at about 250 bp and 170 bp were thought as wild-type and homozygous mouse, respectively. While heterozygotes had both two bands.

Mouse grouping and administration

After genotyping, mice were kept to 12 months. HdhQ140 mouse appeared less activity, locomotor deficient, and gait abnormal at 12 months. All mice were operated behavioral experiments grouped by gene type and the abnormal mice were excluded at 12 months. Having finished the pilot test, the formal protocol was conducted. J3 was dissolved and distributed in corn oil to 12 mg or 24 mg/mL, mice were intraperitoneal injection of two day’s dosage of J3 suspension 0.1 mL per 20 g and 3 times per week.

Mouse behavioral experiments

In the behavioral experiment, the gene types and grouping message of all mice were blind and all experiments were conducted in the light phase. Both male and female mice had been used, between two mice to be tested, 75% alcohol was prepared to eliminate substances and smell that left in machine or chamber.

For the activity test, mice were placed in the dark chamber (110 mm × 100 mm × 110 mm) for 5 min, the rearing number indicated the total number of events that the mice lifted up at least two limbs. The counts of rearing numbers were calculated by machine automatically and showed on the screen of the machine.

For open-field tests, mice were placed in a blue chamber (110 mm × 100 mm × 110) mm in the behavioral test room, and locomotion was captured by a camera on top of the chamber and recorded for 15 min. The times of event that the mouse lifting its forelimbs, traveling tract and distance were then analyzed by the behavioral analysis system.

Based on the animal’s nature to explore a novel object more than a familiar object, novel object recognition (NOR) was designed to test the ability of distinguishing objects. Mice were placed in the testing room and stayed for 1 h to reduce stress, experimental animals were briefly handled for 3 days. On day 1, animals were habituated in the empty open field for 30 min to lower basal stress levels. On day 2, animals were presented with two identical cylinders (A1 and A2) in the chamber for 5 min to finish the training session. The cylinders were placed opposite each other 10 cm from each wall. After 24 h, in the NOR test, animals were exposed to a familiar object (A2) and a novel object (B). The novel object was a cube that had a different size to the familiar object. A2 and B were placed in the same locations as A1 and A2 were placed in the training sessions. Object exploration was measured by the stopwatch, and ‘exploration’ was defined as sniffing or touching the object with the nose. Behavior was not scored as ‘exploration’ when the animal was using the object to rear up or when the animal was sitting on the object. A percent of time spent exploring the novel object (TN) relative to the total time spent exploring both objects (TN + TF, TF refers to the time devoted to the familiar object) can be a measure of novel object recognition. Data are expressed as a recognition index (RI): RI = TN/ (TN + TF).

For the balance beam test, mice were pre-trained to across the balance beam that settled at 30-degree incline cradle in one-way. According to the reference, beams could be 80 cm long, 15 mm or 10 mm diameter smooth round one or 15 mm or 10 mm side cube beams. In the training and experimental session, the mice were started at the lower end to climb. At the end of the inclined beam, there is a 20 × 20 cm wooden platform. A 20 × 20x10 cm cage with a 5 × 5 cm door towards the higher end of beams was set to attract mouse walk into the little house. Having trained the mouse to climb the balance beam in single way, the number of slip faults was counted to evaluate the balance and coordination skills.

For the rotarod test, mice were pre-trained on 3 consecutive days on the rotarod rotating at 4 rpm for 2 min. Mice were then tested for 3 days at 4 rpm, 8 rpm or accelerating speed ranging from 4 to 8 rpm within 2 min. Each performance was recorded as the time in seconds spent on the rotating rod until falling off or until the end of the task. Each test including three repetitions with an inter-trial interval of 1 h to reduce stress and fatigue, and the means from these three runs were analyzed for each mouse.

For the climbing test, a vertical 1 cm diameter smooth metal rod was fixed in the platform. In the training session, mice were placed at the top of the rod, once the mice scratched the rod steadily, the mice climb down along the rod autonomously. The time spent on the climbing that mice across between the top and the platform. The scores of motor evaluation of mice were noted. If the mice slide along the rods step by step, count 0 point; if mice slip length less than 20 cm, count 0.5 points; if mice could not scratch firmly to slip along the rod, count 1 point; if mice could not scratch the rod and drop, count 1.5 points. Each test included three repetitions with an inter-trial interval of 1 h to reduce stress and fatigue, and the means from these three runs were analyzed for each mouse.

Immunohistochemistry

The pretreatment of immunohistochemistry is similar to the immunostaining process. The brain tissue was fixed and cut into thick slices followed by blocking by 5% BSA and incubation with primary antibodies HTT (1:100) and DARPP-32 (1:3000) at 4 ℃ overnight. Then the samples were incubated with secondary antibodies for 30 min at room temperature and developed by DAB agent. Images were captured by EVOS microscopes and analyzed blindly by Image J for quantifications.

Blood–brain barrier measurement

The blood–brain barrier of J3 was investigated in 10-week-old mice following a single intravenous injection (IV) at 6 mg/kg. J3 was prepared in 10% DMA + 20% Solutol + 70% (20%SBE-β-CD in Saline) to get the required solution for IV. Plasma and brain collection intervals of post-dose for IV is at 0.5 h. 3 male ICR mice of SPF were used in each group. In brief, blood samples taken via submandibular vein were placed in tubes containing K2-EDTA and centrifuged at 6800 g for 6 min at 2–8 °C within 1 h. Samples were stored frozen at approximately −80 °C. The brain of mice was collected after the animals were euthanized by CO2 inhalation. The ratio of brain-plasma concentration was calculated by Conc. in brain/ Conc. in plasma.

Statistic assay

All data were derived from three independent experiments. Statistical comparisons between two groups were conducted by the unpaired two-tailed student tests. Statistical comparisons among multiple groups were conducted by one-way ANOVA tests, and Kruskal–Wallis test was applied between groups to be compared when performing statistics in GraphPad Prism 8.0. When the p value was less than 0.05, there was a significant difference between groups.