3,4,5-Tricaffeoylquinic acid induces adult neurogenesis and improves deficit of learning and memory in aging model senescence-accelerated prone 8 mice

TCQA improved spatial learning and memory of SAMP8 mice

To evaluate the effect of TCQA on spatial learning and memory in SAMP8 mice, the Morris water maze (MWM) test was employed. MWM is a widely used tool to assess spatial learning and memory in rodents. In the MWM, the animal is placed on a platform concealed under the surface of the water in the pool. The swimming time each mouse takes to swim to the platform indicates how quickly it can recall the location of the platform. To assess the effect of TCQA on spatial learning and memory, we measured the swimming time of the mice to reach the platform. As shown in Figure 1A, the escape latency time of TCQA-treated SAMP8 mice was significantly decreased compared to water-treated SAMP8 mice from 4^th to 7^th day training. On the other hand, there was no difference in the escape latency between SAMR1 mice and TCQA-treated SAMP8 mice.

Figure 1. Effect of ethanol extract of 3,4,5-triCaffeoylquinic acid (TCQA) on the spatial learning and memory as determined by escape latency of senescence-accelerated resistant mouse 1 (SAMR1) mice, senescence-accelerated prone mouse 8 (SAMP8) mice and SAMP8 TCQA-treated group determined by Morris water maze test (A). Effect of ethanol extract of TCQA on the time spent in the target quadrant (B). Effect of ethanol extract of TCQA on numbers of crossings of platform by SAMR1 untreated and SAMP8 treated or untreated mice (C). * P < 0.05, ** P < 0.01 Compared with SAMP8 + water group.

Figure 1B and 1C show that the time spent in the target quadrant and the numbers of time the mouse crossed the platform were significantly lower in the water-treated SAMP8 group compared to both the TCQA-treated SAMP8 group and the water-treated SAMR1 group. However, no difference was observed in the occupancy and crossing of the target quadrant between the TCQA-treated SAMP8 group and the water-treated SAMR1 group.

TCQA stimulates stem cells and increases neurogenesis in the dentate gyrus of SAMP8 mice

To study the effects of TCQA on NPC proliferation, we performed immunohistochemical detection of BrdU+ cells found in the dentate gyrus (DG) and the subventricular zone (SVZ) of SAMR1 mice and TCQA-treated and control SAMP8 mice. We also performed immunohistochemistry for the astrocytic marker GFAP and the neuronal marker NeuN. Cells that co-expressed BrdU and GFAP (BrdU+/GFAP+ cells) were identified as stem cells and cells that expressed BrdU+/NeuN+ were identified as newborn neurons. Both the numbers of BrdU+/GFAP+ cells and BrdU+/NeuN+ cells were significantly increased in TCQA-treated SAMP8 mice compared to untreated SAMP8 mice indicating that TCQA induced activation of NSC and increased the rate of neurogenesis in the anterior DG (Figure 2A–2C). In contrast, there was no significant change in the number of BrdU+/GFAP+ cells in the posterior DG (Figure 2E). TCQA treatment, however, did increase the number of BrdU+/NeuN+ cells in the posterior DG. These results suggest that increased hippocampal neurogenesis could be related to the improvement in learning and memory observed in MWM (Figure 1).

Figure 2. Effect of oral administration of 3,4,5-triCaffeoylquinic acid (TCQA) on anterior (A–C) and posterior (D–F) DG stem cell activation and neurogenesis. SAMP8 mice were administrated with TCQA (5 mg/kg) for 30 days. Photomicrographs show adult mouse brain in coronal sections containing the anterior (A) and posterior (D) DG processed for immunohistochemical detection of proliferating BrdU+ cells (red) and GFAP, a protein expressed by stem cells in the DG (green). Graphs represents the number of BrdU+ cells that co-express GFAP in anterior (B) and posterior (E) DG. Graphs represent the number of BrdU+ cells that co-express the mature neuronal marker NeuN in the anterior (E) and posterior (F) DG. Each bar represents the mean ±SEM * p < 0.05, ** p < 0.01 vs. SAMP8+TCQA group.

The SVZ neurogenic niche is not involved in spatial memory and can exhibit different molecular regulation compared to the DG (Figure 3). Unlike the anterior DG, no significant difference was found in the number of BrdU+/GFAP+ cells in the SVZ among the three groups of mice (Figure 3B). However, the total number of BrdU+ cells in the SVZ was significantly increased in TCQA-treated SAMP8 mice compared to Water-treated SAMR1 mice (Figure 3C).

Figure 3. Effect of oral administration of 3,4,5-tricaffeoylquinic acid (TCQA) on subventricular zone (SVZ) proliferation. SAMP8 mice were administrated with TCQA (5 mg/kg) for 30 days (A) Photomicrograph shows adult mouse brain coronal sections containing the SVZ processed for immunohistochemical detection of proliferating BrdU⁺ cells (red) and GFAP+ (green), an astrocyte marker found in SVZ NSC. (B, C) Graphs represent the number of BrdU+/GFAP+ and BrdU⁺ cells, respectively in the different treatment groups.

TCQA increased cell viability of hNSCs

To evaluate the cytotoxicity of TCQA on hNSCs, cell viability was examined using the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay (Figure 4). The hNSCs were treated with increasing concentrations of TCQA (1, 5, 10, and 20 μM) in differentiation medium for 24, 48, 72, and 96 h. TCQA increased cell viability in a dose-dependent manner at all treatment times compared to corresponding controls (Figure 4). Cells treated with 20 μM of TCQA showed slightly less viability than the cells treated with 10 μM of TCQA at 72 h.

Figure 4. The effect of 3,4,5-triCaffeoylquinic acid (TCQA) on cell viability of human neural stem cells (hNSCs). hNSCs in undifferentiated state (A), induced differentiation for 96 h (B) and induced differentiation and treated with TCQA for 96 h (C). Time after differentiation, hNSCs were treated with TCQA (1, 5, 10, 20 μM) for 24, 48, 72, and 96 h (D). After the treatment, cell viability was measured by MTT assay. Data was set as % of control. Data were presented as mean ± SD. ** P < 0.01 Compared with control group.

TCQA promoted multilineage differentiation of hNSCs

It is known that hNSCs decide their fate within 24 h of induction of differentiation and differentiate into three kinds of cells; neuron, oligodendrocyte, and astrocyte, within three days [11]. We performed immunocytochemistry and western blotting using differentiation markers of three kinds of cells: β3-tubulin for neurons, MBP for oligodendrocytes and GFAP for astrocytes. Before immunocytochemistry and western blotting, we evaluated the cytotoxicity of TCQA (1–20 μM) on hNSCs by MTT assay. TCQA showed increased cell viability in a dose-dependent manner compared to corresponding control cells at 24h–96h (Figure 4). Immunocytochemistry showed that both control and TCQA-treated cells could differentiate into all lineages (Figure 5A–5C). However, TCQA substantially increased protein levels for all three differentiation markers. Compared to control cells, TCQA-treated cells showed significant increase in β3-tubulin expression at 24 h and 96 h of treatment (Figure 5D), and in MBP expression at 48 h and 96 h of treatment (Figure 5E) compared with those of in control cells. TCQA-treated cells also showed the increasing tendency in GFAP expression at 48 h and 96 h of treatment (Figure 5F). Glial differentiation takes place after neuronal differentiation. Furthermore, imaging by confocal laser scanning microscope suggested that TCQA treatment increased neurite length compared to the control condition (Figure 5A). It is known that β3-tubulin is expressed in both cell bodies and neurites (axons and dendrites) [12]. TCQA increased the expression of β3-tubulin in cell bodies at 24 h and 48 h. Then, both control and TCQA-treated cells extended neurites at 72 h. Finally, at 96 h, TCQA-treated cells showed longer neurites similar to axons. These results correspond with the protein expression by western blotting (Figure 5D). Both control and CQA-treated cells showed very low expression of MBP and GFAP compared to the expression of β3-tubulin (Figure 5B and 5C).

Figure 5. The effect of 3,4,5-tricaffeoylquinic acid (TCQA) on fate, protein expression levels of differentiation markers, and cell proliferation of human neural stem cells (hNSCs). Three differentiation markers (β3-tubulin: neuron, myelin basic protein (MBP): oligodendrocyte, glial fibrillary acidic protein (GFAP): astrocyte) were used. hNSCs were treated with differentiation medium with or without 10 μM TCQA. Expression levels of each differentiation marker were observed using confocal microscopy. Immunofluorescence images demonstrating the expression of β3-tubulin (A), MBP (B), and GFAP (C). hNSCs were treated with differentiation medium with or without 10 μM TCQA for 24 - 96 h. The expression level of each differentiation marker was determined by western blotting. Immunopositive bands of β3-tubulin (D), MBP (E), and GFAP (F) were quantified and expressed as a normalized value compared to glyceraldehyde-3-phosphate dehydrogenase (Gapdh). The cell number and cell viability were measured by ViaCount assay (G). * P < 0.05, ** P < 0.01 significance compared with control group.

TCQA increased protein expression levels of all differentiation markers. There are two possibilities for this; first, TCQA promoted differentiation of hNSCs and second, TCQA promoted cell proliferation of hNSCs. To verify these two theories, we performed ViaCount assay to count cell number. hNSCs were treated with or without TCQA in differentiation medium for 24–96 h. In both untreated control and TCQA-treated groups, cell numbers were increased at 48 h compared to 24 h. However, no change was observed in the number of cells after 48h. There was also no significant difference in number of cells between control and TCQA-treated groups at any treatment time (Figure 5G). Additionally, cell viability was not changed at any treatment time in both control and TCQA-treated groups.

TCQA induced G0/G1cell cycle arrest

Cell cycle regulation is closely related to neurogenesis. It is well established that the G0 phase is related to neuronal differentiation [13]. Some G0/G1 cell cycle regulators, such as tumor suppressor protein p53 (p53), induce G0/G1cell cycle arrest and act as transcriptional factors modulating proneural gene expression. To investigate whether TCQA affects hNSCs cell cycle, we performed a cell cycle assay. hNSCs were treated with growth medium or differentiation medium with or without TCQA for 24 h. In undifferentiated hNSCs, TCQA treatment significantly increased the percentage of G0/G1 phase cells (P = 0.02) while S phase cells showed a significant decrease (P = 0.007). Differentiated hNSCs included more cells existing in the G0/G1 phase than the undifferentiated ones. Moreover, in differentiated hNSCs, TCQA increased the ratio of cells existing in G0/G1phase compared to control (Figure 6A and 6B). Next, we considered the expression of cell cycle regulator p53.

Figure 6. The effect of 3,4,5-tricaffeoylquinic acid (TCQA) on cell cycle and phosphorylation of tumor protein p53 in human neural stem cells (hNSCs). Cell cycle was determined by labeling cellular DNA. hNSCs were treated with growth medium or differentiation medium with or without 10 μM TCQA for 24 h. (A) Ratio of each cell cycle phase as a percentage of total cells. (B) The histograms show the cells in G0/G1 (pink peak on left), S (green center peak) and G2/M (blue peak on right). (C) hNSCs were treated with growth medium or differentiation medium with or without 10 μM TCQA for 24 - 96 h. Data was set as % of undifferentiated control. Data were presented as mean ± SD. * P < 0.05, ** P < 0.01 Compared with undifferentiated control cells. # P < 0.05 significance by student’s t test.

Phosphorylation of serine 15 in human p53 is an important regulator of G0/G1 cell cycle arrest. Phosphorylated p53 can activate p21, a cyclin-dependent kinase (CDK) inhibitor 1, or CDK-interacting protein 1 and p21 inhibits G1/S transition by binding directly to cyclin-CDK complexes. We measured phosphorylation of p53 by western blotting. Phosphorylation of p53 was significantly increased in TCQA-treated cells at 24, 48 and 96 h and decreased at 72 h compared to control cells (Figure 6C).

TCQA affected intracellular Ca²⁺ levels, mitochondrial function, and intracellular reactive oxygen species (ROS) levels during early differentiation

Removal of growth factors initiates differentiation by inducing several early events. Firstly, Ca²⁺ is released from the endoplasmic reticulum into the cytoplasm in the absence of growth factors. Elevated Ca²⁺ levels in mitochondria increase oxidative phosphorylation to produce reactive oxygen species (ROS), which in turn activate signaling pathways related to neurogenesis. Thus, Ca²⁺ is considered a crucial initial factor of differentiation after removal of growth factors. We, therefore, investigated whether TCQA affects Ca²⁺ levels. After 1 min of TCQA treatment in undifferentiated cells, intracellular Ca²⁺ levels were significantly increased compared with undifferentiated controls (Figure 7A). TCQA-treated undifferentiated cells also showed slightly higher Ca²⁺ levels compared with undifferentiated control, but not higher than differentiation without TCQA, suggesting that it is the differentiation but not the TCQA that causes increased Ca²⁺. After 3 min of treatment, the intracellular Ca²⁺ levels were decreased in the TCQA treatment group compared with 1 min of TCQA treatment, suggesting that the elevated Ca²⁺ was transported to the mitochondria. After that, the decreased Ca²⁺ levels were gradually restored to baseline (Figure 7A).

Figure 7. The effect of 3,4,5-triCaffeoylquinic acid (TCQA) on intracellular Ca²⁺ levels, mitochondrial function, and intracellular reactive oxygen species (ROS) levels at very early phase of human neural stem cell (hNSCs) differentiation. hNSCs were pre-treated with Fluo4 AM for 30 min followed by treatment with growth medium or differentiation medium with or without TCQA 10 μM for 1–30 min. Time after differentiation and TCQA treatment, intracellular Ca²⁺ level was detected by measurement of fluorescence intensity (A). hNSCs were treated with growth medium or differentiation medium with or without 10 μM TCQA for 30–180 min. TCQA was treated with rhodamine 123 and detected mitochondrial function by measuring the fluorescence intensity (B). Intracellular ROS levels were detected by measuring fluorescence intensity of DCF oxidized by ROS. hNSCs were pre-treated with DCFH-DA for 1 h followed by treatment with growth medium or differentiation medium with or without 10 μM TCQA for 15–180 min (C). Data was set as % of undifferentiated control. Data were presented as mean ± SD. * P < 0.01, ** P < 0.01 Compared with undifferentiated control cells.

After Ca²⁺ is transported to mitochondria, mitochondrial oxidative phosphorylation is activated. Mitochondrial oxidative phosphorylation uses proton concentration gradients between the inner and outer membranes of this organelle as driving power. The mitochondrial membrane potential (MMP) reflects this proton concentration gradient. Therefore, we investigated whether TCQA can affect mitochondrial oxidative phosphorylation by measuring the MMP. hNSCs were treated in a growth medium or a differentiation medium without (control) or with TCQA for 30–180 min. It was previously reported that removal of growth factors could activate mitochondrial oxidative phosphorylation in 30–180 min. After 30 min of the treatment, TCQA significantly increased MMP compared to both undifferentiated and differentiated controls. And, these increased MMP levels were restored to baseline after 60 and 180 min of TCQA treatment (Figure 7B).

TCQA is known to have high antioxidative activity, so there is a possibility that TCQA can reduce cellular oxidative stress by inducing differentiation. hNSCs were treated in a growth medium or a differentiation medium without (control) or with TCQA for 30–180 min. After 180 min of the treatment, TCQA significantly decreased intracellular ROS levels compared with differentiated controls (Figure 7C). However, there was no significant difference in ROS levels between undifferentiated and differentiated controls.

TCQA affects gene expression during differentiation

We observed differences in protein expression of differentiation markers at 24h (Figure 8A–8C). We also found that hNSCs treated with TCQA could differentiate into more committed lineages than control NSCs. Numerous stemness-related genes are responsible for maintaining the undifferentiated state of NSCs. When differentiation is induced by growth factor withdrawal, decreased expression of stemness gene and increased expression of NSC fate-promoting genes can be observed. We, therefore, evaluated the effect of TCQA on global gene expression in hNSCs during differentiation at 24h. We repeated DNA microarray for three times and focused on genes whose expression was consistently altered in all three experiments. We classified genes into four functional groups highly relevant for neurogenesis: actin cytoskeleton (Supplementary Table 1), chromatin remodeling, neuronal development, and cell cycle-related genes (Table 1). Having dual functions, some genes were classified into multiple groups. TCQA increased bone morphogenetic protein (BMP) signaling pathway-related genes. TCQA increased expression of the BMP5 ligand and the BMP receptor, type II (BMPR2) (Figure 8A). Furthermore, TCQA increased SMAD family member 5 (SMAD5) expression, which is activated by BMP signaling pathway and acts as a transcription factor during neuronal differentiation of hNSCs (Figure 6A) [14]. The expression of neuronal differentiation 1 (NEUROD1) is regulated by the BMP-SMAD signaling pathway and TCQA also increased NEUROD1 expression (Figure 8A). TCQA also increased mitogen-activated protein kinase 14 (MAPK14) and mitogen-activated protein kinase kinase 6 (MAP2K6), which are responsible for coding, and for phosphorylation and activation of p38 respectively (Figure 8B). Activated p38 directly phosphorylates p53. Furthermore, activated p53 acts as a transcriptional factor and regulates zinc finger E-box bind 1(ZEB1). ZEB1 is a transcription factor related to neuronal differentiation. We also found that

Figure 8. The effect of 3,4,5-tricaffeoylquinic acid (TCQA) on gene expressions related to bone morphogenetic protein (BMP) signaling pathway. Human neural stem cells (hNSCs) were treated with differentiation medium with or without 10 μM TCQA for 24 h. Genes expressing BMP ligand BMP5, BMP receptor 2 and SMAD5 downstream BMP signaling pathway as well as the neuronal differentiation transcription factor NEUROD1 were increased by TCQA (A). Genes related to p38–p53 signaling pathway regulating G0/G1 cell cycle arrest of hNSCs activated by the BMP signaling pathway were increased by TCQA (B). Genes related to the Cdc42 signaling pathway regulating neurite extension and activated by the BMP signaling pathway were increased by TCQA (C). Data was set as % of undifferentiated control. Data were presented as mean ± SD. ** P < 0.01 Compared with undifferentiated control.

TCQA increased activation of the cell division cycle 42 (Cdc42) signaling pathway that regulates actin cytoskeleton and neurite extension of hNSCs (Figure 8C).

TCQA preparation

We used TCQA for our study as it promotes the highest neuronal ATP production compared with other CQA derivatives [9]. As the extraction and purification/isolation of a large amount of TCQA from plant materials / natural biological sources is difficult, we used synthesized TCQA. Dr. Kozo Sato from Synthetic Organic Chemistry Laboratories, the FUJIFILM Corporation (Kanagawa, Japan) provided us with the synthesized TCQA with 97% purity. TCQA was dissolved in 70% EtOH (TCA stock), and then, for in vitro assays, TCA stock was dissolved in the medium. For in vivo assays, TCA stock was dissolved in drinking water.

Animal subjects, TCQA administration and Bromodeoxyuridine (BrdU) labeling

Four-month-old adult male SAMP8 and SAMR1 (purchased from the Japan SLC company, Japan) mice were used for the experiment. All mice were allowed to acclimate to the laboratory conditions for seven days under controlled conditions of temperature (21–23ºC) and light (light: dark 12:12) with free access to food and water. SAMP8 mice were randomly divided into two groups, SAMP8 control group (n = 6) and TCQA-supplemented group (n = 7). SAMR1 mice (n = 10) were used as normal aging controls. TCQA (5 mg/kg) was mixed with drinking water and then was directly administered by oral gavage with a feeding tube and a syringe every day for 30 days. BrdU (1mg/ml) was diluted in drinking water and was provided to the animals for nine consecutive days starting from the 14^th day of TCQA treatment. The schedule of BrdU administration followed by washout and sacrifice was chosen, as it is capable of detecting differences in stem cell activation as well as neurogenesis. All animal procedures were approved by the Animal Study Committee of University of Tsukuba (No.16–042) and were handled according to the guidelines for the Care and Use of Animals approved by the Council of the Physiological Society of Japan.

Morris Water Maze (MWM)

MWM was carried out using a circular pool (120 cm in diameter and 45 cm in height) with a featureless inner surface. The pool was filled to a depth of 30 cm with water (23 ± 2°C) and was divided into four quadrants. A platform (10 cm in diameter) was placed in the northeast quadrant and was submerged 1 cm below the water surface so that it was invisible at water level. The mice were then given four trial sessions each day for one week on a platform fixed in the same location for the whole duration of the experiment. Each mouse was placed on the platform for 15 s every day for 7 days.

After 7 days of trials, the hidden platform was removed from the pool and the mice were then placed in the opposite location of the platform and were allowed to swim for 60 s. The number of crossings over the previous position of the platform and the time spent in the target quadrant in which the platform was hidden during the acquisition trials were recorded as measures for spatial memory.

Tissue processing and immunostaining

At the end of MWM, animals were sacrificed and brains were removed. Ten brains of SAMR1 mice, six brains of SAMP8 control mice and seven brains of TCQA supplemented SAMP8 mice were harvested. The brains were fixed in 4% paraformaldehyde (PFA) for 24 h and then cryoprotected in 30% sucrose (w/v) in phosphate buffered saline (PBS) for 48 h, both at 4 ºC. Serial 30 µm coronal brain sections were obtained using a microtome on dry ice. Sections were stored at -20ºC in cryoprotectant solution (ethylene glycol, glycerol, 0.1 M phosphate buffer, pH 7.4, 1:1:1:2 by volume). Sections were washed in 0.1 M PBS 3 × 10 min and then blocked with 50 mM glycine in PBS to reduce autofluorescence of PFA-fixed tissue. For immunohistological detection of BrdU, sections were incubated in HCL (2N) at 37ºC for 1 h before blocking. Sections were washed in PBS, before blocking for 1 h in PBS containing 0.1% Triton X-100 and 10% Donkey Serum (DS, Sigma). Primary antibodies used were anti-BrdU (1:400, sheep, Abcam Ab1893), anti-GFAP (1:500, rat, Invitrogen, AF568), anti-NeuN (1:400, mouse, Millipore MAB377). The secondary antibodies were conjugated to Alexa-488, -568 or -647 (Invitrogen Paisley, Renfrewshire, UK; 1:500).

Culture of hNSCs

The hNSCs derived from H9 hNSCs were purchased from Gibco, Japan. The hNSCs were cultured in 60 mm Petri dishes (BD Falcon) or in 6-well plates (BD Falcon) with a growth medium Knock Out^TM:1:1 (v/v) mixture of Dulbecco’s minimum essential medium (Sigma, U.S.A.) and Ham’s F-12 nutrient mixture (Sigma, U.S.A.) supplemented with 2% StemPro neural supplement, 20 ng/mL of fibroblast growth factor (FGF) basic recombinant human, 20 ng/mL epidermal growth factor (EGF) recombinant human and 2 mM GlutaMAX-I supplement. For adhesion, 60-mm Petri dish or 6-well plate coated with CELLStart were used. The hNSCs were cultured at 37ºC in a 95% air / 5% CO₂ humidified incubator. The medium was changed every two days to keep cells undifferentiated.

Determination of cell viability

Cell viability was measured using the MTT assay. hNSCs were treated with 5–30 μM of TCQA for 72 h. After TCQA treatment, 5 mg/ml of MTT was added and the cells were incubated for further 12 h. The MTT formazan was dissolved in 100 μl of 10 % SDS (w/v). Absorbance (570 nm) was measured using a microtiter plate reader (Dainippon Sumitomo Pharma Co., Ltd., Japan).

Differentiation assay and immunocytochemistry

To induce differentiation, hNSCs were seeded into CELLStart coated culture vessels at the density of 2.5 x 10⁴ cells/cm². After 48 h, growth medium was replaced by differentiation medium without growth factors: KnockOut DMEM/F12 supplemented with 2% StemPro neural supplement and 2 mM GlutaMAX-I supplement. The medium was changed every 2 to 3 days. For immunocytochemistry, cells were seeded in the Nunc Lab-Tek Chamber Slide System (Thermo Scientific, Japan) at the same density and were incubated at 37ºC for 48 h. After incubation, cells were treated with differentiation medium with or without TCQA at 37ºC for 24–96 h. After treatment, cells were washed with PBS and were fixed with 4% (w/v) PFA diluted in PBS for 30 minutes at room temperature. The fixed cells were washed with PBS 3 times for 5 minutes. After washing, cells were blocked overnight at 4ºC with saturation solution of 5% Fetal Goat Serum and 0.1% TritonX-100 (Sigma, Japan) in PBS. Blocking buffer was removed after incubation and primary antibodies (Novex Goat anti-Rabbit IgG (H+L), Sigma, Japan) were added in a specific antibody dilution buffer (1% Bovine Serum Albumin and 0.3% TritonX-100 in PBS). After the incubation with the primary antibody overnight at 4ºC, cells were washed with PBS and incubated with the secondary antibody (Alexa Flour 488 conjugate goat anti-rabbit IgG (H+L) (1:1000), Thermo SCIENTIFIC, Japan) for 1–2 h at room temperature avoiding light exposure. After incubation, cells were washed with PBS and nuclei were counterstained with DAPI using drops of ProLong Gold Antifade Mountant (Thermo Scientific, Japan). Fluorescence was detected with Leica DMI 4000B (Leica, Germany).

Protein extraction

The hNSCs were seeded in 60 mm Petri dish (BD Falcon) at the density of 2.5×10⁴ cells/cm² and were incubated at 37ºC for 48 h. After the incubation, cells were treated with differentiation medium with or without 10 μM TCQA for 24–96 h at 37ºC. After the treatment, cells were washed with cold PBS twice and then 100 μl of RIPA buffer (Sigma, Japan) including 1% protease inhibitor (Sigma, Japan) was added. Cells were incubated for 5 min on ice and cell lysate was collected in Cryo-tubes and was frozen in liquid nitrogen for 15 min. The frozen solution was melted on the ice and centrifuged (4ºC, 10,000g, 15 min) in 1.5 ml Protein LoBind Tube (Eppendorf, Japan). The supernatants were transferred into new tubes and protein concentrations were quantified with 2-D Quant kit (GE Healthcare, Japan). Protein samples were stored at -80ºC until use.

Western blotting

Protein samples (20 µg) were separated using 10% SDS-PAGE and were transferred to a polyvinylidene difluoride membrane (PVDF, Merck Millipore, USA). Western blot was performed as previously described (Matsukawa et al., 2017). Primary antibodies used were anti-β-III-Tubulin (CST, Japan); anti-myelin basic protein (MBP) (Sigma, Japan); anti-glial fibrillary acidic protein (GFAP) (Sigma, Japan); anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1:200, Santa Cruz, Japan); rabbit anti-Phospho-p53 (S15, CST, Japan) and rabbit anti-p53 (1:1000, CST, Japan). Secondary antibodies were IRDye 800CW donkey anti-rabbit IgG (1:1000) or IRDye 680LT goat anti-mouse (1:20000) from LI-COR, Inc., Lincoln, NE, USA. The signal was detected using the Odyssey FC Imaging System (LI-COR, Inc, NE, USA).

Cell cycle assay

Guava Cell Cycle Reagent (Millipore, Japan) was used to evaluate the effects of TCQA on cell cycle regulation in undifferentiated and differentiated cells. The hNSCs were cultured in 60 mm Petri dishes at a density of 2.5 × 10⁴ cells/cm² and were incubated at 37ºC for 48 h. Cells were treated with or without TCQA (10 µM) in growth or differentiation medium at 37ºC for 24 h. After incubation, cells were washed and treated with StemPro accutase to detach from the dish and were centrifuged twice at 200 g for 4 min. One ml of cold 70% ethanol was next added and cells vortexed. Cells were fixed at 4 ºC for 1–12 h. Then the cells were centrifuged twice at 450 g for 5 min. After the last centrifugation, Guava Cell Cycle Reagent (Guava Technologies, Japan) was added to cells and was incubated for 30 min at room temperature in the darkness. After incubation, cell cycle was determined using Guava easyCyte (Millipore, Japan).

Measurement of intracellular Ca²⁺ levels in hNSCs

After inducing differentiation of hNSCs into neurons by removal of growth factors, intracellular Ca²⁺ level immediately changes to activate certain signaling pathways. Therefore, intracellular Ca²⁺ levels were determined to evaluate the effect of TCQA at a very early phase of differentiation using Calcium KitⅡ-Fluo 4 (Dojindo, Japan).

hNSCs were seeded at 2.5 × 10⁴ cells/cm² in black clear bottom 96-well plates (Greiner). After 48 h of incubation, cells were pre-treated with 100 μl of Loading buffer (5% Pluronic F-127,250 mmol/l Probenecid and 1 μg/μl Fluo 4 AM in Hanks’ HEPES Buffer) for 30 min. After that, the supernatant was removed and cells were washed with PBS and treated with differentiation medium without (control) or with 10 μM of TCQA. The cells were incubated at 37ºC for 1–180 min in the dark. After the treatment, fluorescence intensity (495 nm/535 nm) was detected using a microplate reader and was calculated as a percentage of control.

Rhodamine 123 uptake measurement

Rhodamine 123 is a fluorescent probe that stains mitochondria directly. In this study, rhodamine 123 was used to evaluate mitochondrial membrane potential (ΔMMP). Once hNSCs were induced to differentiate, intracellular Ca²⁺ was released from the endoplasmic reticulum and subsequently, Ca²⁺ was transported to mitochondria. Because mitochondrial Ca²⁺ levels modulate ATP and ROS production, rhodamine 123 was used for evaluating mitochondrial activity. Rhodamine 123 (Wako, Japan) was dissolved in DMSO at 100 mg/ml. hNSCs were seeded at 2.5 × 10⁴ cells/cm² in 96-well plates (BD Falcon) and incubated at 37ºC for 48 h. After the incubation, hNSCs were treated in a growth medium or a differentiation medium without (control) or with 10 μM of TCQA and were incubated at 37ºC for 30–180 min. After the treatment, hNSCs were washed with PBS and were treated with rhodamine 123 (10 μg/ml), dissolved in 10 mM HEPES-HBSS buffer (pH 7.4), and incubated at 37ºC for 20 min. After the incubation, hNSCs were washed with cold-PBS, lysed with 1% Triton-X solution (200 μl/well) and incubated at room temperature for 30 min avoiding light exposure. After the incubation, the cell lysate was transferred into a black clear bottom 96-well plate and the fluorescence intensity (507 nm/529 nm) was measured. Uptake levels of rhodamine 123 were calculated as a percentage of control.

Measurement of intracellular ROS level

Removal of growth factors to induce differentiation of hNSCs causes an elevation of the intracellular Ca²⁺ level, which in turn activates ROS. Under normal physiological conditions, ROS activate certain signaling pathways, however, excess ROS can induce oxidative stress that harms cells. Therefore, the effects of TCQA on intracellular ROS level was determined using the OxiSelect Intracellular ROS assay kit (Cosmo Bio, Japan) following the manufacturer’s procedures.

hNSCs (2.5 × 10⁴ cells/cm²) were seeded in black clear bottom 96-well plates and were incubated in medium with TCQA for 1 h. The cells were then incubated with 10 μM DCFH-DA at 37°C for 1 h and washed with PBS. The fluorescence intensity of dichlorofluoresein (DCF) was measured using a microtiter plate (Dainippon Sumitomo Pharma Co, Ltd., Japan) reader at excitation wavelength of 485 nm and emission wavelength of 528 nm. Intracellular ROS was monitored using the fluorescent probe DCFH-DA. Intracellular H₂O₂ or low-molecular-weight peroxides oxidize DCFH-DA to the highly fluorescent compound DCF.

RNA isolation from hNSCs

The hNSCs were seeded in 60-mm Petri dishes at the density of 2.5×10⁴ cells/cm² and were incubated at 37ºC for 48 h. After that, cells were treated with differentiation medium with or without 10 μM of TCQA and were incubated at 37ºC for 6–24 h. After the treatment, cells were washed with ice-cold PBS. The total RNA was extracted from the cells using ISOGEN kit (Nippon Gene Co. Ltd., Toyama, Japan) following manufacturer’s instructions, as reported previously (Han et al., 2010). Total RNA was quantified and assessed for quality with a NanoDrop 2000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA).

DNA microarray analysis

Double-stranded cDNA was synthesized from 100 ng of total RNA with the GeneAtlas 3’ IVT Express Kit (Affymetrix Inc., Santa Clara, CA, USA). Biotin-labeled amplified RNA (aRNA) was synthesized by in vitro transcription using the GeneChip 3’ IVT Express Kit (Affymetrix Inc., Santa Clara, CA, USA). A total of 9.4 mg of purified aRNA was fragmented using the GeneAtlas 3’ IVT Express Kit and was hybridized for 16 h at 45°C using GeneChip MG-430 PM microarray (Affymetrix Inc., Santa Clara, CA, USA). The chip was washed and stained in the Gene Atlas Fluidics Station 400 (Affymetrix Inc., Santa Clara, CA, USA) and then the resulting image was scanned using the GeneAtlas Imaging Station (Affymetrix Inc., Santa Clara, CA, USA). Data analysis was performed using the Partek Express software (Partek Inc., St. Louis, MO, USA) provided by Affymetrix as part of their GeneAtlas system. Compared with the non-treated cells, fold change in expression in the TCQA-treated group was calculated and converted to log 2 data.

Statistical analysis

All analyses were performed on blind-coded slides as previously described [47]. Statistical analysis was performed using GraphPad Prism 6 (GraphPad Software, Inc., San Diego, CA). A Student’s t-test was used when two groups were compared. Results are expressed as mean ± standard error of the mean (SEM). Statistical analysis of the results obtained in the MWM was carried out using two-way ANOVA with Ryan-Einot-Gabriel-Welsch multiple range test. One-way ANOVA followed by Ryan-Einot-Gabriel-Welsch multiple range test was also done. A P value of < 0.05 was considered statistically significant.