Animals

Except for the 5xFAD studies, all experiments were performed using Sprague-Dawley (SD) rats at 8 to 12 weeks of age (Charles River Japan, Tokyo, Japan). The rats were group housed with 2–3 per cage in a climate-controlled room with a temperature of 23 °C ± 3 °C, and a 12-h dark-light cycle, with food and water ad libitum. These experiments were conducted in compliance with the Guidelines for Animal Experimentation of Japan Tobacco Biological/ Pharmacological Research Laboratories.

For AD model mice studies, 5xFAD mice (APPSwFlLon, PSEN1*M146L*L286V) 6799Vas/Mmjax) and control B6SJL mice were purchased from Jackson Laboratories and maintained by the UCLA Division of Laboratory Animal Medicine. Mice were housed in groups of 5, under a 12-h light/12-h dark cycle at 21 °C, with food and water ad libitum. All the experiments were reviewed and carried out in accordance with National Research Council Guide for the Care and Use of Laboratory Animals and approved by the University of California, Los Angeles Institutional Animal Care Use Committee, and performed with strict adherence to the guidelines set out in the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Inhibitory activity of compound A

For assessing the inhibitory activity of Compound A on human PDHK (hPDHK) and rat PDHK (rPDHK), recombinant hPDHK1-4, rPDHK1-4 with a Flag-tag at the N-terminus and PDH enzyme derived from porcine heart were used. PDHK activity was assessed indirectly by measuring the residual PDH activity after PDHK reaction as described previously [10]. Briefly, PDH (P7032, Sigma-Aldrich, STL, USA) and human/ rat PDHK1, 2, 3 or 4 (Central Pharmaceutical Research Institute, Japan Tobacco Inc.) were mixed and incubated in a refrigerator at 4 °C overnight to obtain a PDH/PDHK complex. The PDH/PDHK complex and Compound A were incubated for 45 min at room temperature in a 96-well half area microplate after adding 0.3 µM ATP (for PDHK1, 2 and 3) or 10 µM ATP (for PDHK4) to start the PDHK reaction. Then, the substrates for PDH in reaction buffer (50 mM MOPS (pH 7.0), 20 mM K2HPO4, 60 mM KCl, 2 mM MgCl2, 0.4 mM EDTA, 0.2% Pluronic F-68, 2 mM DTT, 5 mM CoA, 5 mM Sodium pyruvate, 5 mM TPP and 12 mM β-nicotinamide adenine dinucleotide) were added to start the PDH reaction and incubated for 90 min at room temperature. The absorbance at 340 nm before and after the PDH reaction was measured to determine the production of NADH by the PDH reaction. Three independent trials were performed and the IC50 value of Compound A was calculated for each type of PDHK enzyme. The PDHK inhibition (%) of the test compound, a reversible inhibitor, was calculated from the formula [ × 100]. The IC50 value was calculated from the concentration of the test compound at the two points between the PDHK inhibition rates of 50%.

Rat primary cultures and drug treatment

Primary cortical neurons were prepared from embryonic day 16 SD (IGS) rat embryos (Charles River Laboratories Japan). Briefly, cortical fragments of 8 litters were dissociated into single cells with a Neuron Dissociation Solutions kit (Cat No. 291-78001, FUJIFILM Wako Pure Chemical Corporation, Japan) and resuspended in Neurobasal medium (Cat No. 21103-049, Thermo Fisher Scientific, USA) supplemented with 2% B27 (Cat No. 17504-044, Thermo Fisher Scientific, USA). The neurons were seeded onto poly-L-lysine-coated 96-well plates (Cat No. 356516, Corning, NY 14831 USA) at a density of 80,000 cells per well and cultured in Neurobasal medium supplemented with 2% B27 for 10 days to perform the assay in a 5% CO2/95% air atmosphere. On Day 10, half of the culture medium was removed and collected into a new 96-well plate. The new medium containing the DMSO or Compound A at double the final concentration was added onto both plates, and the plates without cells were designated as pre-treatment media samples. After the cells were incubated for 24 h, the medium was collected as the post-treatment samples.

Primary astrocyte cultures were prepared from the cortex of 3 pups of SD (IGS) rats at postnatal day 1. Briefly, the cells were isolated with a Neuron Dissociation Solutions kit, as described above, and then cultured in DMEM (Cat No. 10569-010, Thermo Fisher Scientific, USA) with 10% heat-inactivated FBS (Cat No. 16000-077,Thermo Fisher Scientific, USA) and 1% N-2 (Cat No. 17502-048, Thermo Fisher Scientific, USA) under 5% CO2/95% air. After the cells reached confluence, they were shaken for 16 h at 200 rpm to remove the microglial cells and oligodendrocytes. After that, astrocytes were removed and cryopreserved at − 80 °C as a stock. Before the assays, cryopreserved astrocytes were rapidly thawed for plating in DMEM with 10% heat-inactivated FBS and 1% N-2 medium and transferred to noncoated flasks and then cultured for 3 days. After the cells had grown, astrocytes were replated on poly-L-lysine -coated 96-well plates at a density of 2,000 viable cells per well. Four days after plating, the medium was replaced with medium containing 250 μm dibutyryl-cAMP (Bucladesine sodium; Cat No. 029-16383, FUJIFILM Wako Pure Chemical Corporation, Japan) and cultured for 15 days before the assay. On Day 20, the medium containing DMSO or Compound A was added to the astrocyte-seeded plates. After the cells were incubated for 24 h, the medium was collected as the post-treatment samples. The culture medium containing DMSO or Compound A was used for the pre-treatment samples.

Quantification of pyruvate, lactate and glucose in the culture medium

The levels of pyruvate, glucose and lactate in the pre- and post-treatment samples were measured by biochemical methods. The pyruvate and lactate were measured using Determiner PA and Determiner LA kits (Minaris Medical, Japan), respectively. Glucose was measured using a Glucose C II-Test Wako kit. The changes of these metabolites in 24 h were calculated as the values of post-treatment samples minus that of the pre-treatment samples.

Brain PDH activity and lactate levels in CSF of rats

The day before administration, sixteen rats were assigned to four groups so that no bias occurred in the body weight between the groups. The vehicle or Compound A was orally administered to each of the rats. 4 h after administration, blood samples were collected from the tail vein while awake and CSF was collected under anesthesia with 2.5% isoflurane. After waking up from anesthesia, whole brain was collected immediately after decapitation, followed by fast freezing with a freezing-clamp.

Brain PDH activity was determined by the AAT-coupled method following previously reported methods [10]. Briefly, the brains were collected in 4 vol of the homogenization buffer and immediately homogenized with the homogenizer. The homogenate was centrifuged at 10,000 g for 20 min and 10 µL of the supernatants were suspended in 150 µL of the sample buffer to obtain active PDH samples. Twenty microliters of the supernatants were suspended with 20 µL of the activation buffer containing 10 mM of dichloroacetate (DCA, Cat No. 347795, Sigma-Aldrich, USA) and incubated at 37°C for 10 min to completely activate the PDH. Then 20 µL of the incubated samples were diluted with 60 µL of the sample buffer to obtain total PDH samples. To eliminate endogenous acetyl-CoA in the samples, 50 µL of each sample was suspended with 130 µL of the reaction buffer and incubated at 37°C for 15 min. After incubation, 20 µL of 50 mM sodium pyruvate solution was added, and conversion of 4-aminoazobenzene-4’-sulfonic acid sodium salt to acetyl-4-aminoazobenzene-4’-sulfonic acid sodium salt was monitored at 460 nm after incubating for 30 min at 30 °C. Brain PDH activity (% active) was expressed as the ratio between the activity of the active PDH samples and total PDH samples, which was then normalized by protein concentration determined with a BCA protein assay kit.

2-DG uptake assay

Vehicle or Compound A was orally administered approximately 16 h before 2-Deoxy-d-glucose (2-DG) administration. Animals were decapitated 2 min after the intravenous injection of 2-DG at a dose of 3 mg/kg. Then the brain was removed, and the right cerebral cortex was dissected. The cortices were homogenized and subsequently heated at 95 °C for 15 min. Following centrifugation (17,360 x g for 15 min at 4 °C) to remove the cell debris, the resulting supernatants were used to measure 2-DG levels. The measurement of 2-DG uptake in the tissue was performed using the 2-deoxyglucose uptake measurement kit (Cosmo Bio, Tokyo, Japan) according to the manufacturer’s instructions, enabling the 2-DG uptake into the cells to be quantified as the amount of 2-deoxyglucose 6-phosphate (2-DG6P) without the interference of the remaining extracellular 2-DG based on the enzymatic photometric method [11].

Efficacy study using 5xFAD AD model mice

The effect of Compound A on cognitive function was evaluated using Y-maze, Novel object recognition test (NORT) and Morris water maze (MWM) with 5xFAD mice. We used twenty-three Non-Tg mice and forty-five 5xFAD mice for behavioral assessment, which were assigned into three groups: Non-Tg-vehicle (n = 23), 5xFAD-vehicle (n = 23) and 5xFAD-Compound A (n = 22), including both males and females with approximately equivalent numbers. They were divided into three cohorts and were subject to behavioral testing in a staggered manner. They were treated with vehicle or Compound A mixed into chow (PicoLab Rodent Diet 5053, Inotiv, Indiana, U.S.A.) (Compound A mixed chow contained Compound A at the dose of 30 mg/kg/day) for three months beginning from seven-months of age, when amyloid deposition was well-established. In all behavioral assessments, mice exhibiting abnormal behaviors that precluded testing such as rotations or sickness-like states such as freezing were excluded. After all behavioral testing was completed, animals were anesthetized with isoflurane and transcardially perfused with phosphate-buffered saline (PBS) to perform biochemical analysis using one hemisphere and the other hemisphere was immersion fixed for immunohistochemistry.

The effect of Compound A on neurodegeneration and Aβ pathology was evaluated using an additional cohort. Four Non-Tg mice and seven 5xFAD mice were divided into the same three groups as above. The mice were treated with vehicle or Compound A with food for one month. After the treatment, animals were processed with the same procedure as above. The detailed procedure is described below.

Behavioral assessmentY-maze

One month after starting treatment with Compound A, the mice were tested with the Y-maze. 21 Non-Tg-vehicle, 19 5xFAD-vehicle and 22 5xFAD-Compound A mice completed testing and were analyzed after excluding mice disqualified due to climbing the walls of the apparatus rather than exploring the maze arms or exhibiting sickness-like immobility. Briefly, they were placed in one of the three arms of the apparatus (7.5 × 38 × 12.5 cm) located in a dim room and were allowed to freely explore three arms for 8 min. All sessions were video recorded. Alternation behavior was defined as consecutive entries into each of all three arms without repeated entries, as on overlapping triplet sets. The analysis was performed in a blinded manner by an independent co-worker based on the recorded videos. Alternation rates were calculated as the ratio of alternation behavior to the (total arm entries – 2) × 100.

NORT

One month after starting treatment with Compound A, the mice were tested with the NORT. The numbers of mice in each group successfully completing NORT and analyzed was 15 Non-Tg vehicle, 12 5xFAD-vehicle and 14 5xFAD-Compound A. The lower n’s completing reflects failure of recording of one of the three cohorts. The NORT was undertaken using a 40 × 40 × 40 cm open field arena, constructed from white plastic material. A LEGO block and a 25 cm2 cell culture flask filled with water colored by black paint were used as the objects. Objects were placed at an equal distance from the arena walls. The placement of the objects was counterbalanced by side and the order of object choice in different sessions was randomly assigned across animals to avoid bias. The arena and all the objects used were cleaned with 10% alcohol between sessions. The NORT consists of 3 days. On day 1, the mice were acclimated to the experimental apparatus in the absence of objects for 10 min. On day 2, they were again allowed to explore the apparatus where the same objects were placed for 10 min. On day 3, the mice were placed in the arena with the same object as the previous day and allowed to explore for 10 min (acquisition trial) and returned to the home cage. 1 h after the acquisition trial, one of the objects were replaced with the novel object, and then the mice were placed and allowed to explore for 10 min (retention trial). Behavior was recorded by a digital camera, and the exploration behavior of the animals was manually analyzed based on the recorded video in a blinded manner. The novelty discrimination index (NDI) was defined as the novel object exploration time divided by the total exploration time in the retention trial.

Morris water maze

Three months after starting treatment with Compound A, the mice were tested with the MWM. 23 Non-Tg vehicle, 15 5xFAD-vehicle mice and 17 5xFAD-Compound A mice completed the MWM. Excluded mice were unable to complete the test due to severe motor deficits. In MWM test, a circular pool with a diameter of 120 cm, a height of 45 cm and a water depth of 25 cm was used, and a colorless transparent platform with a diameter of 12 cm was placed in position about 1 cm below the water surface. All animals were allowed to swim freely in the pool for 60 s to memorize the location of the platform. A 60 s/trial was performed three times/day for five days in the hidden platform test, and a probe test was performed about 4 h after the last trial on the fifth day. In the probe trial, the pool was divided into four quadrants without setting the platform, and the target area was defined as the quadrant where the platform was placed during the hidden platform test, the percentages of time spent in the target area in 60 s were calculated.

Immunohistochemistry

Mice were anesthetized and transcardially perfused with phosphate-buffered saline (PBS). The right hemisphere was fixed in 4% paraformaldehyde and cryoprotected by immersion in 30% sucrose for 3 days at 4˚C for immunohistochemistry. The serial coronal sections (10 μm) obtained with a cryostat were washed three times in PBS, blocked with 5% bovine serum albumin (BSA) and 0.1% Triton X-100 in PBS for 1 h at room temperature. The primary antibodies include: anti-NeuN (1:100, Cat No. ab177487, Abcam) and anti-amyloid beta (6E10) (1:500, Cat No. 803001, Biolegend) at 4˚C overnight. After washing, slices were incubated with secondary antibodies: anti-rabbit 488 (1:3,000, Cat No. A-11004, invitrogen) and anti-mouse 594 (1:3,000, Cat No. A-11005, invitrogen) at room temperature for 2 h. Images were acquired using an FV-3000 Laser-scanning confocal microscope (Olympus).

Quantitative analysis of NeuN + cells was performed on three sections per animal and analyzed in a blinded fashion by ImageJ software (National Institutes of Health (NIH), Bethesda, MD, USA). Analysis was done in the following steps: (1) images are converted to 8 bit; (2) following converting, selected automatic thresholding method for area of NeuN + cells and background signals were removed; (3) the thresholded images for the designated brain area were quantified using Olympus software and the Analyze particles tool for the “count” value of the NeuN + cells; (4) “count” value of the NeuN + cells was then divided by the designated area. Finally, (5), to normalize relative to the control, the following equations are applied: % of control = (Non-Tg or 5xFAD vehicle or treatment/ average of Non-Tg) x 100. The steps for the quantitative analysis of Aβ were the same up to (2). In (3), the thresholded images for the designated brain area were quantified using the Analyze particles tool for the “%Area” value of the Aβ; (4) to normalize relative to the control, the following equations were applied to the “%Area” values of the two groups to be compared: % of 5xFAD = (%Area 5xFAD-vehicle or 5xFAD-Compound A/ %Area average of 5xFAD) × 100.

Metabolomic analysis using liquid chromatography-mass spectrometry (LC-MS)

The frozen brain stored at -80 °C was extracted by adding methanol on ice (added 100 µL/10 mg sample) followed by vortexing for 1 min. The homogenates were centrifuged for 5 min (10,000 rpm, 4 °C), and the supernatant was collected. A mixture of 45 µL of supernatant and 5 µL of internal standard solution (Metabolite Yeast Extract (Cat No. ISO1, Cambridge Isotope Laboratories), L-(+)-Lactic Acid-13C3 Sodium Salt (Cat No. L113507, Toronto Research Chemicals Inc.) and Acetylcholine chloride (Cat No. A6625-25G, Sigma)) was used along with brain methanol extracts for metabolomic analysis.

LC-MS analysis was performed using a Nexera UHPLC/HPLC system and a triple quadrupole mass spectrometer LCMS-8060 (Shimadzu, Kyoto, Japan). The LC system was equipped with a ZIC-pHILIC column (2.1 × 150 mm, 3 μm, Sequant, Darmstadt, Germany). Chromatographic separation was based on the previous report [12]. For mobile phases A and B, 20 mmol/L ammonium bicarbonate in water and acetonitrile were used, respectively. The flow rate was 0.25 mL/min. The sample cooler and column oven temperatures were set at 4 °C and 45 °C, respectively. The gradient of mobile phase B concentration was programmed as 90% (0 min) − 90% (5 min) − 30% (20 min) − 30% (23 min) − 90% (24 min) − 90% (35 min). Sample injection volume was 5 µL. The parameters for the mass spectrometer were set as follows: nitrogen was used for nebulizer gas, drying gas, and heating gas. The flow rate of the nebulizer-, drying-, and heating gas were set at 3 L/min, 10 L/min, and 10 L/min, respectively. Argon was used for collision-induced dissociation (CID). The heat block, desolvation line, and interface temperatures were set at 400 °C, 250 °C, and 300 °C, respectively. Mass spectrometric data was acquired by multiple reaction monitoring (MRM). MRM transitions were based on the software method package ver.1 for the simultaneous analysis of primary metabolites (Shimadzu, Kyoto, Japan).

Traverse MS ver. 1.2.9 (Reifycs, Tokyo, Japan) was used for the data processing, including peak detection and integration, and calculation of peak areas. The peak area of each metabolite was divided by that of corresponding internal standard to calculate relative levels of metabolites. Heatmap visualization of the z-scored levels of metabolites in cerebral cortex was performed using the ggplot2 package using the R software.

Statistical analysis

The results were expressed as means ± standard deviations. Data generated from the above studies were statistically analyzed with GraphPad Prism (version 6.07). Comparisons between the two data sets were performed by unpaired t-test or one-way ANOVA, followed by Dunnett-test. Correlation analysis was performed by linear-regression analysis followed by Pearson test. A P value of < 0.05 was considered statistically significant.