Heme Oxygenase 1 Inhibits Adult Neural Stem Cells Proliferation and Survival via Modulation of Wnt/β-Catenin Signaling

Abstract

Background:

Adult hippocampal neurogenesis is critical for renewing hippocampal neural circuits and maintaining hippocampal cognitive function and is closely associated with age-related neurodegenerative diseases. Heme oxygenase 1 (HO-1) is a stress protein that catalyzes the degradation of heme into free iron, biliverdin, and carbon monoxide. Elevated HO-1 level constitutes a pathological feature of Alzheimer’s disease, Parkinson’s disease, and many other age-related neurodegenerative diseases.

Objective:

Here we research the precise role of HO-1 in adult hippocampal neurogenesis.

Methods:

To explore the effect of HO-1 overexpression on adult neural stem cells (aNSCs) and elucidate its mechanisms, Tg(HO-1) was constructed. The transgenic mice and aNSCs were subjected to neurosphereing assay, clonal analysis, and BrdU labelling to detect the proliferation and self-renewal ability. LiCl, MG132, CHX, and IGF-1 treatment were used to research the signaling pathways which regulated by HO-1.

Results:

HO-1 overexpression decreased proliferation ability and induced apoptosis of aNSCs in subgranular zoon (SGZ) in vivo and in vitro. Furthermore, HO-1 overexpression inactivated canonical WNT/β-catenin pathway. Re-activate canonical WNT/β-catenin pathway rescued aNSCs proliferation and survival upon HO-1 overexpression. More importantly, phosphorylation of AKTS473 and GSK3βS9 was found to be significantly decreased in HO-1 overexpressed aNSCs. Re-activation of AKT signaling proved that HO-1 inhibited Wnt/β-catenin signaling pathway via AKT/GSK3β signaling pathway.

Conclusion:

These results demonstrated a critical role of HO-1 in regulating aNSCs survival and proliferation by inhibiting Wnt/β-catenin pathway through repression of AKT/GSK3β, which provide a novel insight into the role of HO-1 in Alzheimer’s disease pathogenesis.

Keywords

Adult hippocampal neurogenesis age-related neurodegenerative diseases canonical Wnt/β-catenin pathway heme oxygenase 1 neural stem cell

INTRODUCTION

Heme oxygenase 1 (HO-1) is almost undetectable in most cell types in the heathy brain under physiological conditions. However, normally low expressed HO-1 increases dramatically under pathological stimulus in neurons, astrocytes, and microglia of Alzheimer’s disease (AD) [1], Parkinson’s disease [2], and multiple sclerosis [3] which occurs mostly in the elderly. In 1995, Schipper et al. showed a strong overexpression of HO-1 in AD patients’ brains via immunoblot and immunohistochemical analysis of postmortem samples. The elevated levels of HO-1 were co-localized with both neurofibrillary tangle proteins and amyloid plaque proteins [4]. Amyloid precursor protein transgenic mice were found to overexpress HO-1 in amyloid aggregates as senile plaques in AD brains [5]. Some researchers believe this dose-dependent elevation of HO-1 is an attempt to protect the brain from oxidant injury [6]. However, there are many studies revealed a cytotoxic role of HO-1 in AD, such as abnormal iron deposits in astrocytes [7], neural damage [8], and uptake of excitotoxicity neurotransmitter from the synaptic cleft [9], leading to cognitive decline, olfactory damage, and neuroendocrine disorders. Thus, the precise function of HO-1 in AD has to be studied further.

Neural stem cells act as undifferentiated precursor source that have the capacity to proliferate and differentiate into neuronal and glial lineages. Along with the findings of neurogenesis in adult brain, the adult neural stem cells were discovered, which are restricted to two regions in adult mammalian brain: the subgranular zone (SGZ) of hippocampal dentate gyrus and the subventricular zone (SVZ) of lateral ventricles [10]. The proliferation ability, which is a hallmark of stem cells, maintains the stem cells pool throughout life. In the adult brain, the microenvironment provides many signals to regulate the maintenance, self-renewal, proliferation and fate commitment of local stem cells. Illuminating of molecules and signaling pathways in stem cell niche helps us to completely understand the characteristics of adult neural stem cells, to increase opportunities to regulate neurogenesis for therapy [11], as well as tissue engineering [12].

Wnt proteins and signaling have been shown very important for the stemness of stem cells [13], including the hematopoietic stem cells [14, 15], cancer stem cells [16], epidermal/gut progenitors [17], and others. To activate canonical Wnt signaling pathways, the secreted glycoprotein Wnt activates a transmembrane receptor, Frizzled (Frz), which leads to activation of the intracellular protein Dishevelled (Dvl) and subsequent phosphorylation of an inhibitory serine residue (Ser9) on GSK-3β. Deactivation of GSK-3 results in the accumulation of dephosphorylatedβ-catenin[18], which allows β-catenin to stabilize, accumulate in the cytoplasm [19]. The accumulated β-catenin trans-locates to the nucleus to bind to the lymphoid enhancer binding factor (LEF) and T-cell transcription factor (TCF) family of transcription factors, leading to the activation of their target genes expression, such as c-Myc and cyclin D1 [20].

The hippocampal cognitive decline is one of the main characters of AD. The adult neural stem cells (aNSCs) that exists in the hippocampal SGZ region in adult mammalian is responsible for the renewal and maintenance of the hippocampal circuitry, which is essential for hippocampal cognitive function. HO-1 is a sensor for oxidative stress and overexpression of HO-1 was proved induce cognitive decline in mice and promote the pathological features of AD-like oligomers formation in hippocampus [21]. In this article, we show that HO-1 overexpression inhibits neural stem cells (NSCs) proliferation and induces NSCs apoptosis in vivo and in vitro. Our surprising findings provide a novel insight into how HO-1 regulates NSCs function. And thus HO-1 plays a critical role in the pathogenesis of AD.

MATERIALS AND METHODS

Reagents and antibodies

All reagents and antibodies used in this investigation were obtained from commercial sources.

Reagents

TrypLE Express Enzyme (1X), no phenol red (12604013), FGF-B (PMG0035), EGF (PMG8045), B-27 Supplement minus vitamin A (12587010), N-2 Supplement (17502048), L-Glutamine (25030081), NE-PER™ Nuclear and Cytoplasmic Extraction Reagents (78835) were purchased from Gibco, USA; Cycloheximide (sc-3508) was purchased from Santa Cruz Biotechnology, USA; MG132 (HY-13259) was purchased from MedChemExpress, USA; In Situ Cell Death Detection Kit, Fluorescein, TUNEL-FITC (11684795910) was purchased from Roche Applied Science, Germany; IGF-1 (250-19) was purchased from Peprotech, USA.

Primary antibodies

Anti-Heme Oxygenase 1 (ab13248), Anti-Caspase-3 (ab44976), Anti-NeuN (ab177487), Anti-BrdU (ab6326), Anti-pan-AKT (ab8805), Anti-p38 (ab31828), Anti-phospho-p38 (phospho T180) (ab178867), Anti-c-Myc (ab32072)antibodies were purchased from abcam, UK; p-GSK-3-beta (9322), GFAP (12389), Sox2 (4900), β-Catenin (8480), GSK-3β (12456), Cleaved Caspase-3 (9664), bcl-2 (15071), bax (14796), β-actin (4970), β3-Tubulin (5568), phospho-Akt (Ser473) (4046), Cyclin D1 (2922) antibodies were purchased from Cell Signaling Technology, USA; BrdU (B5002) was purchased from Sigma Aldrich, USA; Nestin Antibody (NB100-1604) was purchased from Novus Biologicals, UK.

Secondary antibodies

Alexa Fluor 488-conjugated goat anti-rabbit (ZF-0511), Alexa Fluor 594-conjugated goat anti-rabbit (ZF-0516), Rhodamine (TRITC)-conjugated rabbit anti-goat (ZF-0317), FITC conjugated rabbit anti- goat (ZF-0314), Alexa Fluor 488-conjugated goat anti-mouse (ZF-0512), Alexa Fluor 594-conjugated goat anti-mouse (ZF-0513), Fluorescein(FITC)-conjugated goat anti-rat (ZF-0315), Rhodamine (TRITC)-conjugated goat anti-rat (ZF-0318) were purchased from ZSGB-BIO, China.

Animals

Oxidative stress is gradually increased during aging and is more significantly higher in the aged. In this study, we utilized a transgenic mouse systematically overexpressing HO-1 to imitate the elevated level of HO-1 in various tissues and cells in an aging state.

The generation of HO-1 transgenic mice (C57BL/6 background) has been described previously [8]. Briefly, the cDNA of mouse HO-1 under the control of the chicken β-actin promoter was cloned into pCAGG plasmid. Then linearized fragment was microinjected into the fertilized ova by standard pronuclear injection technique. For in vivo experiments, each group contains 3–5 mice and each experiment is repeated at least 3 times independently. For in vitro experiments, aNSCs of each group were extracted from 3–5 mice and each experiment is repeated at least 3 times independently. All animal experiments were conducted in accordance with the guideline approved by the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Ethics Committee of Harbin Medical University (HMUIRB20150023).

Isolation of adult neural stem cells

In all cases, brain tissues were removed from 2–3-month-old male and female Tg(HO-1) mice or wild type mice (WT). The isolation of adult neural stem cells methods was in accordance with Weixiang Guo’s protocol [22]. Briefly, the brain was placed onto the tissue chopper and chopped into 400μm coronal sections. Then, collected the sections containing hippocampus (∼5 sections) and SGZ regions was dissected out from the hippocampus under a dissecting microscope. After digestion and centrifugation, the cells were seeded into a culture plate and kept in the CO2 incubator for 48 h. Neural spheres formed in 1–2 weeks after changing half the culture medium every other day.

Neural stem cells culture, differentiation, and proliferation analysis in vitro

Neural stem cells were cultured in DMEM/F12 medium containing 1 mM L-Glutamine and N2 supplement, B-27 plus 20 ng/ml epidermal growth factor (EGF), and 20 ng/ml basic fibroblast growth factor-2 (FGF-2). For stem cells passage, neurospheres were harvested after 7–14 days, spun down at room temperature (RT) at 200 g for 5 min, the supernatant was removed and digested in 0.025% trypsin-EDTA for 1 min. Neutralized digestion with the medium and pipetted up and down. After centrifugation and resuspension, the cells were seeded into plates for culturing. For differentiation, isolated neural stem cells were induced with DMEM/F12 medium supplemented with 1 mM l-glutamine, B27, and 0.5% fetal bovine serum for 7–10 days. For proliferation analysis, neural stem cells were treated with 2.5μMBrdU for 2 h and stained with anti-BrdU antibody staining. Proliferation was assessed by neurosphere assay and clonal analysis. For the neurosphereing assay, low-density mouse NSCs (2×105/ml cells) was seeded into plates to form neurospheres. For clonal analysis, single NSC was plated per well in 96-well plates and the cells were monitored continuously for 7–10 days.

LiCl, cycloheximide (CHX), and MG132 treatment

For LiCl treatment, 0 mM, 5 mM, or 10 mM LiCl were applied to neural stem cells for 24 h then subjected to western blot, QRT-PCR, MTT assay, or BrdU staining. For CHX and MG132 treatment, 100μg/ml CHX (translational inhibitor to inhibit protein biosynthesis) or 10μM MG132 (the proteasome inhibitor) were applied to neural stem cells for 0 h, 2 h, 4 h, and 6 h, respectively, and then subjected to western blot. To activate the Akt signaling pathway, aNSCs were treated with IGF-1 (100 ng/ml) for 7days and the cells were subjected to neurosphereing assay, clonal analysis, and BrdU labelling.

BrdUinjection

To assess the number of aNSCs in adult mice, intraperitoneal injection of BrdU (30 mg/kg, BrdU was dissolved in 0.06 N NaOH in 0.9% NaCl and neutralized by HCl before use) was given every 2 h for six times. Then, mice were sacrificed 24 h after last injection and brains were isolated. To determine long-term BrdU retention in SGZ and tracking new neuron formation lineage, BrdU was dissolved in phosphate-buffered saline (1×PBS) and injected (100 mg/kg, i.p.) into 2–3-month-old HO-1 transgenic mice and WT mice once a day for 6 consecutive days. 24 days after last injection, brains were harvested and sectioned. Each first slice of every 10 slices was chosen to be stained.

BrdU assay

For BrdU staining in vitro, aNSCs were treated with 2.5μMBrdU for 2 h and stained with anti-BrdU antibody staining. For BrdU staining in vivo, after BrdU injection, brains were harvested and sectioned, sections were pre-treated with 1M HCl on ice for 30 min, then with 2 M HCl for 10 min at RT, then with 2 M HCl for 20 min in 37°C, and lastly with 0.1 M borate buffer at RT for 10 min. Sections were then blocked with 5% BSA and incubated overnight at 4°C with indicated primary antibodies. The following day, all sections were washed with PBST, and incubated with secondary antibodies for 1h. After three washes, the sections were stained with DAPI and mounted on glass slides. Stained sections were imaged using a Nikon C2 confocal microscope.

TdT-mediated dUTP nick end labelling (TUNEL) assay

TUNEL assay was performed to examine apoptosis of neural stem cells. Specifically, NSCs were fixed with 4% formaldehyde and permeabilized with Cytonin. Streptavidin–fluorescein conjugate was applied to detect the biotinylated nucleotides. Fluorescein-stained cells were analyzed by fluorescence microscopy.

Immunostaining and quantification

The brain tissues were freshly embedded in OCT and sectioned along the entire sagittal extent of the brain using a microtome (CM1950, Leica, Germany). The frozen tissues were cut into 20μm and sections were stored in –80° C. Each first slice of every 10 slices was chosen to be stained. Sections were rinsed three times in PBS for 5 min each and blocked for 1 h. with 5% BSA at RT. Sections were then incubated overnight at 4°C with the indicated primary antibodies diluted in PBST. The following day, sections were rinsed three times in PBS for 5 min each and incubated with the suitable secondary antibodies for 1 h at RT. After three washes, the sections were stained with DAPI and mounted on glass slides. Stained sections were imaged using a Nikon C2 confocal microscope. The total number of BrdU+, BrdU+ NeuN+ cells and area volume estimation in hippocampus were estimated by unbiased stereological method [23]. Briefly, the hippocampus is sectioned in the horizontal plane to 20μm and stained with specific marker, then captured in a raster pattern with 2μm stepping motors with confocal microscope. After sampling systematically at all levels, the specific cell numbers were calculatedin every unbiased counting frame according to the formula provided. For specific area volume estimation, for example, the dentate gyrus, the boundaries of the specific area wasdetermined and the corresponding slices were extracted and calculated according to the formula provided. Then the specific cell numbers were estimated using the following formula: N = ∑Q × 1/ -fraction × numberofseries, where N is equal to the specific cell numbers, ∑Q is the specific cell numbers counted per brain, the fraction is the percentage of the total volume that was used for sampling.

Neural stem cells extracted from HO-1 transgenic mice or WT mice were fixed in 4% paraformaldehyde (PFA) for 20 min at RT, permeabilized with 0.5% Triton X-100 for 10 min, and blocked with 5% BSA for 1 h at RT. Then, cells were incubated with primary antibodies overnight at 4°C and labeled with secondary antibody for 1 h at RT. DAPI was used to stain the nuclei. Immunoreactive cells were detected using a Nikon C2 confocal microscope.

Real-time polymerase chain reaction

Total RNA was isolated from aNSCs of Tg(HO-1) and WT mice with Trizol reagent (Thermo Fisher Scientific, USA) and the complementary DNA (cDNA) was synthesized from 0.5μg total RNA using High Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific, USA) according to the manufacturer’s instructions. Real-time PCR were performed using Power SYBR^® Green PCR Master Mix (Life Technologies, USA). The primers were as follow:

HO-1 Forward: 5′-GAAGAACTTTCAGAAGGGTCAG-3′,

HO-1 Reverse: 5′-TCGTCGGAGACGCTTTACATAG-3′.

β-catenin Forward: 5′-ATGGAGCCGGACAGAAAAGC-3′,

β-catenin Reverse: 5′-CTTGCCACTCAGGGAAGGA-3′.

Cyclin-D1 Forward: 5′-GCGTACCCTGACACCAATCTC-3′,

Cyclin-D1 Reverse: 5′-CTCCTCTTCGCACTTCTGCTC-3′.

c-Myc Forward: 5′-ATGCCCCTCAACGTGAACTTC-3′,

c-Myc Reverse: 5′-CGCAACATAGGATGGAGAGCA-3′.

bcl2 Forward: 5′-GACTGACACTGAGTTTGGCTACG-3′,

bcl2 Reverse: 5′-GAGTCCTTTCCACTTCGTCCTG-3′.

Wnt3a Forward: 5′-ATTGAATTTGGAGGAATGGT-3′,

Wnt3a Reverse: 5′-CTTGAAGTACGTGTAACGTG-3′.

GAPDH-F: 5′-AGCCTCCCGCTTCGCTCTCT-3′,

GAPDH-R: 5′-GCGCCCAATACGACCAAATCCGT-3′.

GAPDH was used for normalization.

Nuclear protein extraction

The nuclear protein of aNSCs from Tg(HO-1) and WT mice was extracted with CelLytic™ NuCLEAR™ Extraction Kit (Cat#: NXTRACT, Sigma, USA) according to the manufacturer’s instructions.

Western blot

Neural stem cells or tissues were lysed in ice-cold RIPA lysis buffer containing protease inhibitor cocktail (Roche, Gemany). Equal amounts of protein samples were separated with SDS-PAGE and transferred to nitrocellulose membranes (Pall Corporation, USA). Membranes were blocked with 5% skim milk for 1 h at RT and incubated with specific primary antibodies overnight at 4° C. After washing and incubating with rabbit or mouse secondary antibodies (Cell Signaling Technology), the blots were visualized by ECL reagent (GE Healthcare, USA). Then the protein bands were analyzed using ImageJ software.

Statistical analysis

Data were obtained from at least three independent experiments and presented as the mean±standard deviation. Differences in comparisons were analyzed using Student’s t test and three or more group means by one- or two-way ANOVA and the significance was determined at *p < 0.05.

RESULTS

Adult hippocampal neurogenesis is decreased in Tg(HO-1) mice

To investigate the effect of HO-1 overexpression on aNSCs, HO-1 transgenic mice (Tg(HO-1) mice) were generated [8]. As shown in Fig. 1A, HO-1 was significantly increased in hippocampus of Tg(HO-1) mice compared to WT mice. No marked reduction in size, volume, and maximum length of the SGZ was observed in 2–3-month-old Tg(HO-1) mice compared with that in WT mice (Fig. 1B–D). We first measured the proliferative rate of aNSCs by short-term in vivo BrdU labeling in 2–3-month-old Tg(HO-1) and WT mice (Fig. 1E). Brain sections encompassing the entire hippocampus were performed with Anti-BrdU staining. Representative figures are showed in Fig. 1F. The results showed that WT mice contained more BrdU positive cells compared to Tg(HO-1) mice and the number of BrdU positive cells in Tg(HO-1) mice was decreased by 21.37% as quantified in Fig. 1G, indicating the proliferation ability was declined in Tg(HO-1) mice (Fig. 1G, Supplementary Table 1). We then measured the hippocampal neurogenesis of aNSCs by long-term in vivo BrdU labeling in 2–3-month-old Tg(HO-1) and WT mice (Fig. 1E) and hippocampal neurogenesis was affected in Tg(HO-1) mice, with a significant decrease in the number of newborn neurons, as shown by immunostaining for BrdU and NeuN in Fig. 1H and quantified in Fig. 1I. Taken together, these data suggested that HO-1 overexpression impaired hippocampal neurogenesis of aNSCs in vivo.

Fig.1

The hippocampal neurogenesis of aNSCs was decreased in Tg(HO-1) mice. A) The protein level of HO-1 in the hippocampus of Tg(HO-1) mice and WT mice. n = 3 for each group. B) DAPI staining (blue) showed the dentate gyrus of WT and Tg(HO-1) mice. C) Quantification the dentate gyrus volume. NS, not significant, p = 0.610. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group. D) The maximum length of SGZ of WT and Tg(HO-1) mice. p = 0.585, Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group. E-a) Schematic diagram of short-term BrdU injection.E-b) Schematic diagram of long-term BrdU injection. F) Immunofluorescence staining for BrdU (green) in adult SGZ of Tg(HO-1) and WT mice. DAPI (blue). G) Quantification of BrdU in Tg(HO-1) and control SGZ. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group. H) Immunofluorescence staining for BrdU (red) and NeuN (green) in adult SGZ of Tg(HO-1) and WT mice. I) Quantification of double positive BrdU/NeuN cells (H) in Tg(HO-1) and control. Data are means±SEM of 3 independent experiments. **p < 0.01, significantly different from control, Student’s t-test. These experiments were independently repeated at least 3 times.

HO-1 overexpression decreases proliferation ability of aNSCs in vitro

The primary aNSCs were isolated from the SGZ region of 2–3-month-old Tg(HO-1) mice and WT mice respectively (Fig. 2A, B), and identified by co-staining of Nestin and SOX2, two common used neural stem cell markers (Fig. 2C). The differentiation ability of isolated aNSCs was assessed by a neuronal marker Tuj1 and an astrocyte marker GFAP (Fig. 2D). To examine the effect of HO-1 on aNSCs proliferation ability in vitro, primary aNSCs from adult Tg(HO-1) mice and WT mice were cultured for 7 days to allow formation of neurospheres (Fig. 2E). The quantity and size of primary neurospheres were significantly reduced in Tg(HO-1) mice compared with that in WT mice (Fig. 2E, F). Single aNSC was plated per well in 96-well plates and monitored continuously for 9 days for clonal analysis (Fig. 2G). As shown in Fig. 2H, the diameter of neurosphere clones from Tg(HO-1) cells was significantly reduced compared with that of WT clones. And the cells viability was also decreased in aNSCs isolated from Tg(HO-1) mice (Fig. 2I). Consistently with in vivo results, BrdU labeling showed a significant reduction in the number of proliferative aNSCs from Tg(HO-1) (Fig. 2J, K). These data suggested that HO-1 overexpression decreases proliferation ability of aNSCs in vitro.

Fig.2

The proliferation ability of aNSCs isolated from Tg(HO-1) mice was decreased. A) Architecture of the SGZ niche from which we extracted aNSCs. a) Schema of the whole mouse brain chopped into 400μm sections to dissect SGZ region. b) Schema of coronal section showing the hippocampus, subgranular zone (SGZ), subventricular zone (SVZ). B) Level of HO-1 in primary aNSCs from WT and Tg(HO-1) mice. C) Immunofluorescence staining showed that isolated cells were SOX2 (red) and Nestin (green) double positive cells. D) Cells were immunostained for Tuj1 (red) and GFAP (green). E) Neurosphereing assay was performed using aNSCs extracted from WT and Tg(HO-1) mice. F) Quantification of primary neurosphere numbers in (E). Data are means±SEM of 3 independent experiments. n = 5 for each group. *p < 0.05, significantly different from control, Student’s t-test. G) Clonal analysis assay was performed using aNSCs extracted from WT and Tg(HO-1) mice. H) Relative diameter of neurospheres formed in (G). Data are means±SEM of 3 independent experiments. **p < 0.01, significantly different from control, Student’s t-test. n = 5 for each group. I) Cells viability was assessed by MTT. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, 2-way ANOVA. n = 5 for each group. J) SOX2 (red) and BrdU (green) immunostaining in aNSCs from WT and Tg(HO-1) mice. K) Quantification of SOX2 + BrdU+ cells. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 5 for each group. These experiments were independently repeated at least 3 times.

HO-1 overexpression induces apoptosis of aNSCs in SGZ region

To determine if the reduction number of aNSCs is partly due to an increase in apoptosis, apoptosis occurring in aNSCs in vivo were assessed by immunofluorescence staining of cleaved caspase3 (C-Caspase3) and Nestin. Surprisingly, C-Caspase3 + Nestin+ cells which indicated as apoptotic aNSCs were significantly increased in Tg(HO-1) mice compared with those in WT mice (Fig. 3A, B). Similar to our observations in vivo, the number of apoptotic aNSCs in vitro labeled with Tunel + SOX2+ was significantly increased from Tg(HO-1) mice (Fig. 3C, D) compared to those from WT mice. In addition, the protein levels of total caspase3 and cleaved caspase3 were significantly increased upon HO-1 overexpression, indicating an increased apoptosis in aNSCs in Tg(HO-1) mice (Fig. 3E, F). Moreover, the level of anti-apoptotic protein bcl-2 was decreased, and the level of pro-apoptotic protein bax was increased in aNSCs from Tg(HO-1) mice compared with those from WT mice (Fig. 3G, H). These data suggested that HO-1 overexpression induces apoptosis in aNSCs.

Fig.3

HO-1 overexpression induced apoptosis of aNSCs in vivo and in vitro. A) Immunofluorescence staining of C-Caspase3 (red) and Nestin (green) in the SGZ of WT and Tg(HO-1) mice. arrow: apoptosis aNSCs. B) Quantification of percentage of C-Caspase3 + Nestin+ cells in Nestin+ cells in the SGZ of WT and Tg(HO-1) mice. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group (C) TUNEL assay were performed to show aNSCs apoptosis. D) Quantification of percentage of TUNEL + SOX2+ cells in SOX2+ cells. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group. E) Level of HO-1, Caspase3 and C-Caspase3 in primary aNSCs from WT and Tg(HO-1) mice. F) Quantification of relative level of HO-1, Caspase3 and C-Caspase3 in (E). Data are means±SEM of 3 independent experiments. *p < 0.05, ***p < 0.001 significantly different from control, Student’s t-test. n = 3 for each group. G) Level of HO-1, bax and bcl-2 in primary aNSCs from WT and Tg(HO-1) mice. H) Quantification of relative level of HO-1, bax and bcl-2 in (G). Data are means±SEM of 3 independent experiments. ***p < 0.001 significantly different from control, Student’s t-test. n = 5 for each group. These experiments were independently repeated at least 3 times.

HO-1 overexpression antagonizes canonical Wnt/β-catenin signaling

To investigate the molecular mechanism how HO-1 regulated these processes, we examined the expression of β-Catenin, a key effector of the canonical WNT/β-Catenin signaling pathway in aNSCs in vivo and in vitro. Slices of WT mice and Tg(HO-1) mice were co-stained with Nestin and β-catenin. We found that level of β-Catenin was suppressed in Tg(HO-1) mice compared with that in WT mice (Fig. 4A). The same result was obtained in primary cultured aNSCs from Tg(HO-1) mice and WT mice (Fig. 4B, C). But by real-time PCR, no significant difference was observed in β-Catenin mRNA level between aNSCs from Tg(HO-1) mice and WT mice, indicating that HO-1 did not affect β-Catenin transcription in aNSCs (Fig. 4D). And MG132 treatment to inhibit proteasome function found that HO-1 overexpression did not alter β-Catenin accumulation rate in aNSCs (Fig. 4E, F). These data indicated that HO-1 overexpression did not affect β-Catenin production. To examine the β-Catenin degradation, CHX were applied to inhibit protein biosynthesis. We found that the level of β-Catenin was decreased more sharply in aNSC from Tg(HO-1) mice compared with that from WT mice, indicating that HO-1 overexpression facilitated β-Catenin degradation in aNSCs (Fig. 4G, H). What’s more, we found that HO-1 overexpressed aNSCs harbored relative lower β-catenin expression and less β-catenin localized to nucleus in aNSCs compared to WT aNSCs (Fig. 5A, B), indicating that HO-1 inhibited the translocation of β-catenin into the nucleus. GSK3β is important for β-Catenin phosphorylation and subsequently ubiquitination and degradation. Indeed, higher total level of GSK3β, p-Ser9- GSK3β and lower level of p- Tyr216-GSK3β were observed in aNSC from Tg(HO-1) mice compared with that from WT mice (Fig. 5C, D). These data demonstrate that HO-1 overexpression increased the activity of GSK3β, promoted degradation of β-catenin, and thus inhibited the translocation of β-catenin into the nucleus, leading to the inactivation of canonical Wnt/β-catenin signaling.

Fig.4

HO-1 overexpression inhibited WNT/β-catenin signaling pathway. A) Immunofluorescence staining of Nestin (green) and β-catenin (red) in brains of WT and Tg(HO-1) mice. B) Western blot showed the protein level of β-catenin in primary aNSCs from WT and Tg(HO-1) mice. C) Quantification of β-catenin levels in (B). Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group. D) qRT-PCR was used to determine the mRNA level of β-catenin. p = 0.314, Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group. E) β-catenin levels upon MG132 treatment assessed by western blot. F) Quantification of β-catenin levels in (E). p = 0.007, p = 0.045, Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, 2-way ANOVA. n = 3 for each group. G) β-catenin levels upon CHX treatment assessed by western blot. H) Quantification of β-catenin levels in (G). Data are means±SEM of 3 independent experiments. *p < 0.05, **p < 0.01, significantly different from control, 2-way ANOVA. Data are means±SEM of 3 independent experiments. **p < 0.01, ***p < 0.0001, significantly different from control, Student’s t-test. n = 3 for each group. These experiments were independently repeated at least 3 times.

Fig.5

HO-1 overexpression inhibited WNT/β-catenin signaling pathway. A) Immunofluorescence staining of β-catenin (green) and DAPI in primary aNSCs from WT and Tg(HO-1) mice. B) Expression of β-catenin in nuclear. n = 3 for each group. C) Levels of β-catenin, GSK3β, p-Ser9-GSK3β and p-Tyr216 GSK3β in primary aNSCs from WT and Tg(HO-1) mice. D) Quantification of levels of HO-1, f β-catenin, GSK3β, p-Ser9-GSK3β and p-Tyr216 GSK3β in (C). Data are means±SEM of 3 independent experiments. **p < 0.01, ***p < 0.0001, significantly different from control, Student’s t-test. n = 3 for each group. These experiments were independently repeated at least 3 times.

Re-activation of Wnt/β-catenin signaling pathway rescues HO-1 induced stemness features decline in aNSC

To explore whether re-activation of Wnt/β-catenin signaling could restore stemness features in aNSC upon HO-1 overexpression, aNSCs from Tg(HO-1) mice and WT mice were treated with LiCl, an inhibitor of GSK3β that activates Wnt/β-catenin signaling pathway. We found that LiCl administration significantly restored the expression of β-catenin in aNSCs overexpressed of HO-1 (Fig. 6A–D). And the protein levels of GSK3β, p-GSK3β, and β-Catenin, the components of WNT/β-Catenin signaling pathway, and cMyc, CyclinD1, and bcl2, the downstream target of WNT/β-Catenin signaling pathway, were significantly increased in the primary aNSCs from Tg(HO-1) mice brain in response to LiCl treatment (Fig. 6C, D). The mRNA levels of these target genes showed the same pattern (Fig. 6E). We have proved in Fig. 5A and 5B that HO-1 overexpressed aNSCs harbored relative lower β-catenin expression and less β-catenin localized to nucleus in aNSCs compared to WT aNSCs. However, LiCl remarkably accumulated β-catenin level and nuclei localization compared to PBS treatment in aNSCs overexpressed of HO-1 (Fig. 6F–H). These data indicated that LiCl treatment re-activation of Wnt/β-catenin signaling pathway in aNSCs in Tg(HO-1) mice.

Fig.6

LiCl re-activated WNT/β-catenin signaling pathway in aNSCs overexpressed HO-1. A) 0mM, 5mM, and 10mM LiCl were applied to aNSCs extracted from WT or Tg(HO-1) mice. β-catenin level was analyzed by western blot. B) Quantification of relative expression of β-catenin in (A). Data are means±SEM of 3 independent experiments. *p< 0.05, **p < 0.01, significantly different from control, 2-way ANOVA. n = 3 for each group. C) Levels of β-catenin, p-GSK3β, GSK3β, cMyc, Cyclin D1, and bcl2 upon LiCl treatment. β-actin were served for normalization. D) Quantification of protein levels in (C). Data are means±SEM of 3 independent experiments. *p < 0.05, **p < 0.01, significantly different from control, Student’s t-test. n = 3 for each group. E) mRNA expression of cMyc, Cyclin D1, and bcl2 upon LiCl treatment. n = 3 for each group. F) LiCl treatment restored β-catenin expression and nuclear localization in Tg(HO-1) aNSCs. G) Expression of β-catenin in nuclear. H) Quantification of protein levels in (G), Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 3 for each group. These experiments were independently repeated at least 3 times.

Later, we found that the proliferation capacity of HO-1 overexpressed primary aNSCs was restored determined by larger neurospheres formation (Fig. 7A, B). LiCl treatment notably reversed stem cells frequency in aNSCs overexpressed of HO-1 (Fig. 7C, D). In addition, cell viability was restored as well by LiCl treatment in HO-1 overexpressed aNSCs (Fig. 7E). And the number of SOX2 and BrdU double positive cells, which represented proliferative aNSCs, was significantly increased by LiCl treatment upon HO-1 overexpression (Fig. 7F, G).

Fig.7

Re-activation of Wnt/β-catenin signaling pathway rescues HO-1 induced stemness features decline in aNSCs. A) Clonal analysis assay was performed using aNSCs from WT and Tg(HO-1) mice upon LiCl treatment. B) The proliferation capacity of aNSCs was determined by relative diameter of neurospheres formed in (A). Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. n = 5 for each group. C) Neurosphereing assay was performed using aNSCs from WT and Tg(HO-1) mice and applied to PBS or LiCl, low-density mouse aNSCs (2×105/ml cells) were plated into plate to form neurospheres. D) The stem cell frequency was assessed by the number of primary neurospheres. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, Student’s t-test. E) The stem cells viability was assessed by MTT. Data are means±SEM of 3 independent experiments. **p < 0.01, significantly different from control, 2-way ANOVA. n = 3 for each group. F) Immunofluorescence staining of BrdU (green) and SOX2 (red) in primary Tg(HO-1) aNSCs or WT aNSCs. G) Quantification of percentage of BrdU + SOX2+ cells in SOX2+ cells in (F). Data are means±SEM of 3 independent experiments. **p < 0.01, ***p<0.0001, significantly different from control, Student’s t-test. n = 3 for each group. These experiments were independently repeated at least 3 times.

Together, these results demonstrate that re-activated Wnt/β-catenin signaling pathway rescued HO-1 induced stemness features decline in aNSC.

HO-1 overexpression promotes GSK3β activity via AKT signaling pathway

To elucidate the specific pathway by which HO-1 promoted activity of GSK3β, leading to accelerated degradation of β-catenin. GSK3β activity is usually regulated by AKT and p38/MAPK signaling pathway. However, the western blot data showed no significant difference of p38 phosphorylation between aNSC from Tg(HO-1) mice and WT mice, indicating that HO-1 overexpression did not affect p38/MAPK signaling pathway (Fig. 8A, D). Next as shown in Fig. 8A–C, phosphorylation status of AKTS473 and GSK3βS9 was significantly decreased in aNSCs from Tg(HO-1) mice, compared to this from WT mice, indicating that HO-1 regulated GSK3β activity through AKT signaling pathway.

To strength our data, IGF-1 was applied to aNSCs to activate the Akt signaling pathway [25]. Indeed, IGF-1 treatment significantly re-activated AKT signaling upon HO-1 overexpression (Fig. 8E, F). Later we found that IGF-1 treatment restored cell viability in aNSCs overexpressed of HO-1 (Fig. 8G). The stem cells frequency was reversed by IGF-1 treatment, because the quantity of primary neurospheres was significantly increased in Tg(HO-1) + IGF-1 group compared to Tg(HO-1) + PBS group (Supplementary Figure 1A, B). And re-activation of AKT signaling pathway restored the proliferation capacity of HO-1 overexpressed aNSCs determined by larger neurospheres formation (Supplementary Figure 1C, D). Further, SOX2 and BrdU co-immunostaining revealed that IGF-1 treatment significantly increased the number of proliferative aNSCs upon HO-1 overexpression (Fig. 8H, I). These data suggested that HO-1 overexpression promotes GSK3β activity through inhibiting AKT signaling pathway.

Taken all data together, our findings demonstrated that HO-1 suppressed WNT/β-catenin pathway mainly by blocking AKT/GSK3β signaling pathway, subsequently leading to the stemness feature decline as well as reduced cell survival in aNSCs (Fig. 8J).

Fig.8

HO-1 overexpression promotes GSK3β activity via AKT signaling pathway. A) Protein levels in Tg(HO-1) aNSCs or WT aNSCs were analyzed by western blot. Representative blots of p-GSK3βS9, total GSK3β, p-AktS473, total AKT, p-p38 and total p38 are shown. B-D) Protein levels were quantified by densitometry in (A). Data are means±SEM of 3 independent experiments. **p < 0.01, ***p < 0.0001, significantly different from control, Student’s t-test. n = 3 for each group. E) IGF-1 re-activated AKT and increased GSK3β phosphorylation. F) Quantification of protein levels in (E). Data are means±SEM of 3 independent experiments. *p < 0.05, **p < 0.01, ***p < 0.0001, significantly different from control, Student’s t-test. n = 3 for each group. G) The cells viability was assessed by MTT. Data are means±SEM of 3 independent experiments. *p < 0.05, significantly different from control, 2-way ANOVA. n = 3 for each group. H) Immunofluorescence staining of BrdU (green) and SOX2 (red) in primary Tg(HO-1) aNSCs or WT aNSCs upon IGF-1 treatment. I) Quantification of percentage of BrdU + SOX2+ cells in SOX2+ cells in (H). Data are means±SEM of 3 independent experiments. **p < 0.01, ***p< 0.0001, significantly different from control, Student’s t-test. n = 3 for each group. J) Schematic model of HO-1 inhibited Wnt/β-catenin signaling pathway through repressing AKT signaling pathway in aNSCs. n = 3 for each group. These experiments were independently repeated at least 3 times.

DISCUSSION

Hippocampal aNSCs are a small population cells in brains with self-renewal and differentiation capabilities to generate new neurons, astrocytes, and oligodendrocytes to replace the damaged cells or to establish new neural circuits [24]. Dysfunction of hippocampal aNSCs is involved in many neurodegenerative diseases such as AD [25, 26]. In contrast, functions of HO-1 in aNSCs are still poorly understood. Here the first time we reported a crucial role for HO-1overexpression in central nervous system homeostasis, in which HO-1, via inhibition of Wnt/β-catenin signaling pathway by blocking AKT/GSK3β signaling pathway, regulated aNSCs proliferation and survival in adult mouse brain. Our results reveal a previously unrecognized critical role of HO-1 in regulating aNSCs proliferation and survival which provide a new potential mechanism for the decline of adult hippocampal neurogenesis in AD.

Despite a very small population of proliferative aNSCs in adult brains, the functions of these proliferative aNSC cells are critical. The damaged cells will be replaced or new neural circuits will be established by newborn neurons, astrocytes, and oligodendrocytes differentiated from aNSC cells [10]. A small decrease in aNSCs population or in regenerative capability can affect the functions of brains [10]. Studies have shown that nicotinamide phosphoribosyltransferase reduction in proliferative aNSCs resulted in the decrease in the birth rate of DG neurons and the reduction of neurogenesis by conditional KO nicotinamide phosphoribosyltransferase in mice. In our study, overexpression of HO-1 caused approximately 23% reduction in proliferative aNSCs in vivo which is not as much as obtained in vitro experiments. This reduction caused by HO-1 overexpression indeed impaired aNSCs proliferation and survival, characterized by reduced BrdU positive aNSCs and increased level of cleaved caspase-3.

HO-1 is the key enzyme in hemedegradation. Studies have shown that HO-1 is only increased in AD patients with cerebral amyloid angiopathy (CAA), but not in AD patients without CAA [27]. CAA is a vascular lesion present in up to 95% of AD patients and produces MRI-detectable microbleeds in many of these patients which in turn induces HO-1 expression [28]. These observations prompt us the hypothesis that the hemedegradation pathways are probably induced by vascular injury and/or microhemorrhagic changes. Elevated HO-1 degrades heme but eventually causes damage in aNSC in hippocampus.

Upregulated HO-1 in astrocytes promoted pathological brain iron deposition and oxidative mitochondrial, leading to the damage in aging and degenerating neural tissues such as AD [29]. Also, the elevated expression of HO-1 is known to damage the patterns of brain sterol and redox homeostasis [30]. Moreover, HO-1 was proved modulate microRNA expression and mRNA targets, which may contribute to the neural damage in chronic brain disorders [31]. In conclusion, chronic elevation of HO-1 in the stimulated neurons could lead to increased iron deposition [32], oxidative stress [33], mitochondrial damage [34], and macroautophagy [35], which may eventually relate to redox disorder and bioenergy failure in aging-related disease such as AD and Parkinson’s disease. HO-1 may also affect neuronal plasticity and cell survival by regulating cerebral steroid metabolism and neurotoxic protein degradation pathways [36]. We have previously conducted a series of continuous studies between HO-1 and neurodegenerative diseases, and found the pathogenic role of HO-1 in AD. We have demonstrated that long-term overexpression of HO-1 causes cognitive decline, promotes AD-like tau aggregation and synaptic aberrations, induces the aggregation of neurotoxic tau oligomers and Aβ oligomers, and provides a potential pathway for the pathogenesis of tauopathies [37 –39].

The current study proved that HO-1 is involved in the regulation of various tissue regeneration processes, modulating the proliferation, differentiation, survival, and maturation of embryonic stem cells, mesenchymal stem cells, hematopoietic stem cells, and muscle stem cells [40]. However, the role of HO-1 in neurogenesis is poorly understood and direct correlation studies have only reported the effect of HO-1 on the proliferation and survival of neural stem cells in the SVZ region [41]. In conclusion, HO-1 plays an important role in a variety of tissue regeneration processes, affecting different types of stem cells, and HO-1 is also involved in the neurogenesis. However, studies between HO-1 and neurogenesis in the SGZ region are rare.

Many studies demonstrated that HO-1 has a protective effect in stem cells [42] and HO-1 also shows different effects on the differentiation and maturation process of different types of cells [31]. Most of these studies used a functional deletion strategy to make the expression or activity of HO-1 absent. As an inducible heme oxygenase, HO-1 is widely present in various tissues, and its anti-oxidation, anti-inflammatory, and anti-apoptotic effects have been widely accepted. When HO-1 is absent, it will surely affect the process of oxidative stress response in stem cells; it is not difficult to understand the protective effect of HO-1. Notably, HO-1 is also known as “oxidative stress responder” and it is highly susceptible to oxidative stress, inflammation, and other factors and this means in the presence of high levels of oxidative stress such as AD, the HO-1 expression presents a state of long-term upregulated [4]. In this state, the function and mechanism of HO-1 may differ from the conventional concept of protection. Therefore, we used a Tg(HO-1) mice model systematically overexpressed HO-1 to simulate the high expression of HO-1 in multiple organs and multiple types of cells in aging, which is of more practical meaning for chronic degenerative diseases such as AD.

Previous studies have shown that WNT/β-catenin signaling pathway regulated adult hippocampal neurogenesis [43]. And we found that WNT/β-catenin signaling pathway was inactivated in aNSCs in Tg(HO-1) mice determined by significantly decreased level of β-catenin and downstream targets including bcl-2, c-Myc, and Cyclin D1. However, the expression of Wnt3a, an important ligand of canonical Wnt signaling pathway, did not show any difference between aNSCs extracted from Tg(HO-1) mice and WT mice (Supplementary Figure 2). This finding indicated that HO-1did not affect the stimulation of the Wnt pathway.

β-catenin is the effector of WNT/β-catenin signaling pathway. Accumulated β-catenin translocates into cell nuclei and binds to TCF and LEF to induce downstream target genes expression. The level of β-catenin is directly regulated by GSK3β. GSK3β can phosphorylate β-catenin which leads to its subsequent ubiquitination and degradation by proteasome. In aNSCs of Tg(HO-1) mice, GSK3β activity was remarkably increased. And LiCl treatment successfully rescues stemness features decline in aNSC in Tg(HO-1) mice. Thus, we believe that overexpression of HO-1 inactivated Wnt signaling pathway by inhibiting signaling transmission rather than ligands stimulation.

The key regulators in activation or inactivation of GSK3β during neurodevelopment consist of AKT and p38/MAPK signaling pathway [44]. The phosphatidylinositol 3 kinase (PI3K)/Akt pathway was reported elicit a survival signal, which was closely related to HO-1 [45, 46]. Here we found that HO-1 overexpression only represses AKT signaling pathway, rather than p38/MAPK signaling pathway. Re-activation of AKT signaling pathway by IGF-1 treatment successfully re-activates WNT/β-catenin signaling pathway and rescues stemness features decline in aNSC in Tg(HO-1) mice.

Many studies have demonstrated a vital role of neurogenesis in brain of depression patients. Various chronic antidepressant treatments increase adult hippocampal neurogenesis [47], and thereby block or reverse the atrophy and damage caused by stress [48]. Santarelli et al. [49] found that chronic serotonin selective reuptake inhibitors administration stimulates hippocampal stem cell proliferation rates/survival and that disrupting antidepressant-induced neurogenesis blocks behavioral responses to antidepressants. Jayatissa et al. [48] showed a correlation between recovery from anhedonia, as measured by cessation of behavioral deficits in the chronic mild stress rat model of depression and an increase in cytogenesis in the dentate gyrus of the ventral hippocampal formation. Our findings that HO-1 overexpression inhibited proliferation and survival of aNSCs may also provide a new potential function of HO-1 in depression.

Adult hippocampal neurogenesis promotion was proved to improve spatial memory and pattern separation [50], and a decline in neurogenesis may result in cognitive impairments associated with age-related neurodegenerative diseases and disorders such as AD [51]. A growing body of evidence shows that in AD the central molecular regulators that affect the generation of new born hippocampal neurons, and alterations in neurogenesis occurs earlier than the onset of hallmark lesions of disease or neuronal loss [52]. A potential therapeutic strategy to halt or delay age-related neurodegenerative diseases and disorders is to preserve or potentiate the production of new neurons. Moreover, understanding the mechanisms of adult neurogenesis will contribute to reveal insights into the pathogenesis of age-related neurodegenerative diseases and disorders. Our work demonstrates that HO-1 plays a critical role in aNSCs proliferation and survival. It specifically suppresses WNT/β-catenin pathway by blocking AKT/GSK3β signaling pathway. Our investigation is the first to show the role of HO-1 in neural stem cells biology, which suggests HO-1 could play a critical role in AD pathogenesis.

Conclusion

In conclusion, our study provides comprehensive evidence for HO-1 overexpression on adult neural stem cells proliferation ability and survival, and SGZ neurogenesis by suppressingWnt/β-catenin pathway through repression of AKT/GSK3β in mice brain, which suggests HO-1 could play a critical role in the pathogenesis of AD.

Footnotes

ACKNOWLEDGMENTS

This work was supported in part by National Natural Science Foundation of China (81671255, 81570534, and 81773165) and Sub-project of National Science and Technology Major Project (2018YFA0108603).

Authors’ disclosures available online ().

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

References

Smith

, Kutty

, Richey

, Yan

, Stern

, Chader

, Wiggert

, Petersen

, Perry

(1994) Heme oxygenase-1 is associated with the neurofibrillary pathology of Alzheimer’s disease. Am J Pathol 145, 42–47.

Schipper

, Liberman

, Stopa

(1998) Neural heme oxygenase-1 expression in idiopathic Parkinson’s disease. Exp Neurol 150, 60–68.

Mehindate

, Sahlas

, Frankel

, Mawal

, Liberman

, Corcos

, Dion

, Schipper

(2001) Proinflammatory cytokines promote glial heme oxygenase-1 expression and mitochondrial iron deposition: Implications for multiple sclerosis. J Neurochem 77, 1386–1395.

Schipper

, Cisse

, Stopa

(1995) Expression of heme oxygenase-1 in the senescent and Alzheimer-diseased brain. Ann Neurol 37, 758–768.

Premkumar

, Smith

, Richey

, Petersen

, Castellani

, Kutty

, Wiggert

, Perry

, Kalaria

(1995) Induction of heme oxygenase-1 mRNA and protein in neocortex and cerebral vessels in Alzheimer’s disease. J Neurochem 65, 1399–1402.

Jazwa

, Cuadrado

(2010) Targeting heme oxygenase-1 for neuroprotection and neuroinflammation in neurodegenerative diseases. Curr Drug Targets 11, 1517–1531.

Schipper

, Bennett

, Liberman

, Bienias

, Schneider

, Kelly

, Arvanitakis

(2006) Glial heme oxygenase-1 expression in Alzheimer disease and mild cognitive impairment. Neurobiol Aging 27, 252–261.

, Peng

, Hui

, Zhang

, Zhou

, Li

, Si

, Li

, Wang

, Li

, Dong

, Gao

(2015) Overexpression of heme oxygenase 1 impairs cognitive ability and changes the plasticity of the synapse. J Alzheimers Dis 47, 595–608.

Cuadrado

, Rojo

(2008) Heme oxygenase-1 as a therapeutic target in neurodegenerative diseases and brain infections. Curr Pharm Des 14, 429–442.

10.

Bond

, Ming

, Song

(2015) Adult mammalian neural stem cells and neurogenesis: Five decades later. Cell Stem Cell 17, 385–395.

11.

Taupin

, Gage

(2002) Adult neurogenesis and neural stem cells of the central nervous system in mammals. J Neurosci Res 69, 745–749.

12.

Cheng

, Chen

, Chang

, Huang

, Wang

(2013) Neural stem cells encapsulated in a functionalized self-assembling peptide hydrogel for brain tissue engineering. Biomaterials 34, 2005–2016.

13.

Van Camp

, Beckers

, Zegers

, Van Hul

(2014) Wnt signaling and the control of human stem cell fate. Stem Cell Rev 10, 207–229.

14.

Reya

, Duncan

, Ailles

, Domen

, Scherer

, Willert

, Hintz

, Nusse

, Weissman

(2003) A role for Wntsignalling in self-renewal of haematopoietic stem cells. Nature 423, 409–414.

15.

Duncan

, Rattis

, DiMascio

, Congdon

, Pazianos

, Zhao

, Yoon

, Cook

, Willert

, Gaiano

, Reya

(2005) Integration of Notch and Wnt signaling in hematopoietic stem cell maintenance. Nat Immunol 6, 314–322.

16.

Takebe

, Harris

, Warren

, Ivy

(2011) Targeting cancer stem cells by inhibiting Wnt, Notch, and Hedgehog pathways. Nat Rev Clin Oncol 8, 97–106.

17.

, Cheng

, H

TKV

, Lange

, McKinney

, Seidel

, Sanchez Alvarado

(2015) Egr-5 is a post-mitotic regulator of planarian epidermal differentiation. Elife 4, e10501.

18.

Logan

, Nusse

(2004) The Wnt signaling pathway in development and disease. Annu Rev Cell Dev Biol 20, 781–810.

19.

Gordon

, Nusse

(2006) Wnt signaling: Multiple pathways, multiple receptors, and multiple transcription factors. J Biol Chem 281, 22429–22433.

20.

Katoh

, Katoh

(2007) WNT signaling pathway and stem cell signaling network. Clin Cancer Res 13, 4042–4045.

21.

Ewing

, Haber

, Maines

(1992) Normal and heat-induced patterns of expression of heme oxygenase-1 (HSP32) in rat brain: Hyerthermia causes rapid induction of mRNA and protein. J Neurochem 58, 1140–1149.

22.

Guo

, Patzlaff

, Jobe

, Zhao

(2012) Isolation of multipotent neural stem or progenitor cells from both the dentate gyrus and subventricular zone of a single adult mouse. Nat Protoc 7, 2005–2012.

23.

West

, Slomianka

, Gundersen

(1991) Unbiased stereological estimation of the total number of neurons in thesubdivisions of the rat hippocampus using the optical fractionator. Anat Rec 231, 482–497.

24.

Bonaguidi

, Song

, Ming

, Song

(2012) A unifying hypothesis on mammalian neural stem cell properties in the adult hippocampus. Curr Opin Neurobiol 22, 754–761.

25.

Lathrop

, Nanasi

, Varro

(1990) In vitro cardiac models of dog Purkinje fibre triggered and spontaneous electrical activity: Effects of nicorandil. Br J Pharmacol 99, 119–123.

26.

, Gage

(2011) Adult hippocampal neurogenesis and its role in Alzheimer’s disease. Mol Neurodegener 6, 85.

27.

Schrag

, Kirshner

(2016) Neuropsychological effects of cerebral amyloid angiopathy. Curr Neurol Neurosci Rep 16, 76.

28.

Schrag

, Crofton

, Zabel

, Jiffry

, Kirsch

, Dickson

, Mao

, Vinters

, Domaille

, Chang

, Kirsch

(2011) Effect of cerebral amyloid angiopathy on brain iron, copper, and zinc in Alzheimer’s disease. J Alzheimers Dis 24, 137–149.

29.

Song

, Zukor

, Lin

, Liberman

, Tavitian

, Mui

, Vali

, Fillebeen

, Pantopoulos

, Wu

, Guerquin-Kern

, Schipper

(2012) Unregulated brain iron deposition in transgenic mice over-expressing HMOX1 in the astrocytic compartment. J Neurochem 123, 325–336.

30.

Hascalovici

, Song

, Liberman

, Vaya

, Khatib

, Holcroft

, Laferla

, Schipper

(2014) Neural HO-1/sterol interactions {in vivo: Implications for Alzheimer’s disease. Neuroscience 280, 40–49.

31.

Lin

, Song

, Cressatti

, Zukor

, Wang

, Schipper

(2015) Heme oxygenase-1 modulates microRNA expression in cultured astroglia: Implications for chronic brain disorders. Glia 63, 1270–1284.

32.

McCarthy

, Sosa

, Gardeck

, Baez

, Lee

, Wessling-Resnick

(2018) Inflammation-induced iron transport and metabolism by brain microglia. J Biol Chem 293, 7853–7863.

33.

McMahon

, Ding

, Acosta-Jimenez

, Frangova

, Henderson

, Wolf

(2018) Measuring in vivo responses to endogenous and exogenous oxidative stress using a novel haem oxygenase 1 reporter mouse. J Physiol 596, 105–127.

34.

Bansal

, Biswas

, Avadhani

(2014) Mitochondria-targeted heme oxygenase-1 induces oxidative stress and mitochondrial dysfunction in macrophages, kidney fibroblasts and in chronic alcohol hepatotoxicity. Redox Biol 2, 273–283.

35.

Nakamura

, Kageyama

, Yue

, Huang

, Fujii

, Ke

, Sosa

, Reed

, Datta

, Zarrinpar

, Busuttil

, Kupiec-Weglinski

(2018) Heme oxygenase-1 regulates sirtuin-1-autophagy pathway in liver transplantation: From mouse to human. Am J Transplant 18, 1110–1121.

36.

Schipper

, Song

, Tavitian

, Cressatti

(2019) The sinister face of heme oxygenase-1 in brain aging and disease. Prog Neurobiol 172, 40–70.

37.

Hui

, Wang

, Li

, Zhang

, Jin

, Ma

, Zhou

, Nakajima

, Zhao

, Gao

(2011) Long-term overexpression of heme oxygenase 1 promotes tau aggregation in mouse brain by inducing tau phosphorylation. J Alzheimers Dis 26, 299–313.

38.

, Wang

, Zhang

, Wang

, Li

, Peng

, Sun

, Hui

, Gao

(2018) Hemeoxygenase 1 induces tau oligomer formation and synapse aberrations in hippocampal neurons. J Alzheimers Dis 65, 409–419.

39.

Wang

, Hui

, Peng

, Tang

, Jin

, He

, Li

, Zhang

, Li

, Zhou

, Li

, Ma

, Li

, Gao

, Luo

(2015) Overexpression of heme oxygenase 1 causes cognitive decline and affects pathways for tauopathy in mice. J Alzheimers Dis 43, 519–534.

40.

Kozakowska

, Szade

, Dulak

, Jozkowicz

(2014) Role of heme oxygenase-1 in postnatal differentiation of stem cells: A possible cross-talk with microRNAs. Antioxid Redox Signal 20, 1827–1850.

41.

Nada

, Tulsulkar

, Shah

(2014) Heme oxygenase 1-mediated neurogenesis is enhanced by Ginkgo biloba (EGb 761(R)) after permanent ischemic stroke in mice. Mol Neurobiol 49, 945–956.

42.

Cai

, Guo

, Teng

, Nong

, Tan

, Book

, Zhu

, Wang

, Du

, Wu

, Xie

, Hong

, Li

, Bolli

(2015) Preconditioning human cardiac stem cells with anHO-1 inducer exerts beneficial effects after cell transplantation in the infarcted murine heart. Stem Cells 33, 3596–3607.

43.

Lie

, Colamarino

, Song

, Desire

, Mira

, Consiglio

, Lein

, Jessberger

, Lansford

, Dearie

, Gage

(2005) Wntsignalling regulates adult hippocampal neurogenesis. Nature 437, 1370–1375.

44.

Hur

, Zhou

(2010) GSK3 signalling in neural development. Nat Rev Neurosci 11, 539–551.

45.

Rojo

, Salina

, Salazar

, Takahashi

, Suske

, Calvo

, de Sagarra

, Cuadrado

(2006) Regulation of heme oxygenase-1 gene expression through the phosphatidylinositol 3-kinase/PKC-zeta pathway and Sp1. Free Radic Biol Med 41, 247–261.

46.

, Hsu

, Hsieh

, Lin

, Lai

, Wung

(2006) Upregulation of heme oxygenase-1 by Epigallocatechin-3-gallate via the phosphatidylinositol 3-kinase/Akt and ERK pathways. Life Sci 78, 2889–2897.

47.

Morieri

, Rigato

, Frison

, Simioni

, D’Ambrosio

, Tadiotto

, Paccagnella

, Lapolla

, Avogaro

, Fadini

(2020) Effectiveness of dulaglutide vs liraglutide and exenatide once-weekly. A real-world study and meta-analysis of observational studies. Metabolism 106, 154190.

48.

Warner-Schmidt

, Duman

(2006) Hippocampal neurogenesis: Opposing effects of stress and antidepressant treatment. Hippocampus 16, 239–249.

49.

Santarelli

, Saxe

, Gross

, Surget

, Battaglia

, Dulawa

, Weisstaub

, Lee

, Duman

, Arancio

, Belzung

, Hen

(2003) Requirement of hippocampal neurogenesis for the behavioral effects of antidepressants. Science 301, 805–809.

50.

Clelland

, Choi

, Romberg

, Clemenson

Jr. , Fragniere

, Tyers

, Jessberger

, Saksida

, Barker

, Gage

, Bussey

(2009) A functional role for adult hippocampal neurogenesis in spatial pattern separation. Science 325, 210–213.

51.

Kempermann

(2015) Activity dependency and aging in the regulation of adult neurogenesis. Cold Spring Harb Perspect Biol 7, a018929.

52.

Sahay

, Scobie

, Hill

, O’Carroll

, Kheirbek

, Burghardt

, Fenton

, Dranovsky

, Hen

(2011) Increasing adult hippocampal neurogenesis is sufficient to improve pattern separation. Nature 472, 466–470.