Neuroprotective effects of neural stem cells pretreated with neuregulin1β on PC12 cells exposed to oxygen-glucose deprivation/reoxygenation

Abstract

Studies on ischemia/reperfusion (I/R) injury suggest that exogenous neural stem cells (NSCs) are ideal candidates for stem cell therapy reperfusion injury. However, NSCs are difficult to obtain owing to ethical limitations. In addition, the survival, differentiation, and proliferation rates of transplanted exogenous NSCs are low, which limit their clinical application. Our previous study showed that neuregulin1β (NRG1β) alleviated cerebral I/R injury in rats. In this study, we aimed to induce human umbilical cord mesenchymal stem cells into NSCs and investigate the improvement effect and mechanism of NSCs pretreated with 10 nM NRG1β on PC12 cells injured by oxygen-glucose deprivation/reoxygenation (OGD/R). Our results found that 5 and 10 nM NRG1β promoted the generation and proliferation of NSCs. Co-culture of NSCs and PC12 cells under condition of OGD/R showed that pretreatment of NSCs with NRG1β improved the level of reactive oxygen species, malondialdehyde, glutathione, superoxide dismutase, nicotinamide adenine dinucleotide phosphate, and nuclear factor erythroid 2-related factor 2 (Nrf2) and mitochondrial damage in injured PC12 cells; these indexes are related to ferroptosis. Research has reported that p53 and solute carrier family 7 member 11 (SLC7A11) play vital roles in ferroptosis caused by cerebral I/R injury. Our data show that the expression of p53 was increased and the level of glutathione peroxidase 4 (GPX4) was decreased after RNA interference-mediated knockdown of SLC7A11 in PC12 cells, but this change was alleviated after co-culturing NSCs with damaged PC12 cells. These findings suggest that NSCs pretreated with NRG1β exhibited neuroprotective effects on PC12 cells subjected to OGD/R through influencing the level of ferroptosis regulated by p53/SLC7A11/GPX4 pathway.

Introduction

Ischemic stroke is a common type of cerebrovascular disease caused by the blockage of an artery supplying blood to the brain and one of the major causes of death and disability worldwide (Feigin et al., 2016; Campbell et al., 2019; MartInez-Coria et al., 2021; Zhanga et al., 2021). Ischemic stroke limits the supply of oxygen and glucose to the brain, which is also the key cause of neuronal cell death in cerebral infarction. Restoring blood reperfusion as early as possible is crucial for sustaining neuronal viability (Carden and Granger, 2000). However, the reperfusion process triggers additional injury in the ischemic brain (referred to as cerebral ischemia/reperfusion (I/R) injury) that causes irreversible neuronal damage in the brain. Clarifying the mechanism underlying I/R injury and effective prevention strategies has been formidable challenges in stroke treatment (Carden and Granger, 2000; Daubail et al., 2016).

Stem cell therapy has emerged as an important strategy for treatment of cerebral I/R injury, and various stem cell sources have been explored in clinical trials (Hicks and Jolkkonen, 2009; Chen et al., 2022). Among the various stem cell sources, neural stem cells (NSCs) are considered an ideal cell type for the treatment of cerebral I/R injury and have been applied in practice (Liu et al., 2009; Tang et al., 2017). However, the use of NSCs has some limitations, such as the difficulty in obtaining human NSCs because of ethical issues and the low survival, differentiation and proliferation rates of NSCs, which have restricted their clinical application (Kokaia and Darsalia, 2011; Lindvall and Kokaia, 2011; Casarosa et al., 2014). Therefore, an effective method to obtain NSCs/neural progenitor cells (NPCs) and enhance the neurogenic potential of NSCs is urgently required.

The neurotrophic factor neuregulin1β (NRG1β) is a member of the neuregulin family that contains an epidermal growth factor-like motif and play a vital role in the nervous system through activating ErbB receptor tyrosine kinase (Buonanno and Fischbach, 2001). NRG1β regulates the proliferation, differentiation, and migration of neural cells (including NSCs, NPCs, glial cells, and neurons), as well as synaptogenesis and synaptic plasticity (Buonanno and Fischbach, 2001; Birchmeier, 2009). NRG1β is best known for its function in regulating central nervous system injury and repair (Corfas et al., 2004). Administration of NRG1β has been shown to exhibit neuroprotective effects on ischemic stroke in rats (Shyu et al., 2004; Guo et al., 2006; Li et al., 2007). NRG1β treatment also significantly attenuates rat cortical neuron and primary hippocampal neuron damage under oxygen-glucose deprivation (OGD) conditions (Guan et al., 2015; Lu et al., 2016; Yoo et al., 2019). Notably, our previous studies showed that NRG1β exhibited a neuroprotective role in a cerebral I/R rat model by improving the microenvironment of neuron survival, delaying the phase of irreversible neuron necrosis, and inhibiting the mitochondrial apoptosis pathway (Li et al., 2008, 2009; Gu et al., 2017). These reports indicate that NRG1β might be valuable for treating cerebral I/R injury.

Ferroptosis is a newly discovered nonapoptotic cell death characterized by accumulated reactive oxygen species (ROS) and lipid peroxidation and is correlated with the onset and development of various neurological diseases with I/R injury (She et al., 2020). Previous studies reported that p53 plays a vital role in ferroptosis. Activation of p53 inhibites expression of SLC7A11 (a key component of system Xc⁻), reduces cystine uptake and changes glutathione (GSH) synthesis. GSH is required for the activity of glutathione peroxidase 4 (GPX4); when the System Xc-GPX4-GSH-cysteine pathway is inhibited, this process could accelerate ferroptosis (Xie et al., 2016; Chen et al., 2021). Emerging studies have suggested that ferroptosis induces and aggravates brain tissue damage following I/R injury, accompanied by accumulation of ROS, iron transport disorder, and imbalance of the GPX4- GSH-cysteine axis (She et al., 2020; Shen et al., 2020).

In this study, we aimed to explore the potential of NSCs pretreated with NRG1β for the treatment of cerebral I/R injury and examined its regulatory mechanism. Using OGD/reoxygenation (OGD/R) as an in vitro model for cerebral I/R injury, we investigated the effects on NSCs (induced from human umbilical cord mesenchymal stem cells, hUC-MSCs) pretreated with NRG1β on damaged PC12 cells. In addition, the possible mechanism underlying the neuroprotective effects of NSCs pretreated with NRG1β on PC12 cells was explored.

Methods

Isolation and culture of human umbilical cord (hUC)-MSCs

hUC tissues (n = 4) were obtained from healthy full-term pregnant woman in the Affiliated Hospital of Qingdao University who delivered by cesarean section. The samples were immediately cultured in sterile phosphate buffered saline (PBS; Hyclone, Logan, UT, USA, Cat# SH30256.01) containing 1% penicillin-streptomycin (Hyclone, Cat# SV30010). hUC-MSCs were isolated from fresh UC tissues following previously described methods (Mennan et al., 2013; Hong et al., 2020). Briefly, the whole UC was immersed in 75% ethanol for 1 minute and then immediately washed with PBS for three times. The hUC was then cut into large segments (approximately 5 cm in size) and the trapped blood within the umbilical blood vessels was removed. The tissue pieces were washed gently with PBS and the veins and arteries were removed. The hUC segments were further minced into very fine pieces, approximately 0.5–1 cm³ in size, and placed in a 100 mm plate (Corning, New York, NY, USA, Cat# 430167). A drop of Dulbecco’s modified Eagle medium (Gibco, Grand Island, NY, USA, Cat# C11965500BT) supplemented with 10% fetal bovine serum (Gibco, Cat# 10099-141C), 1% penicillin-streptomycin, and 1% L-glutamine solution (Sigma, St. Louis, MO, USA, Cat# G7513) was placed on each segment, and the medium was slowly added over the UC tissue after 8 hours. The UC samples were incubated at 37°C with 5% CO₂, and the medium was changed every 2–3 days. After 10–15 days of culture, the cells that grew densely around the tissue blocks were primary cultured hUC-MSCs. Cells were passaged when cell growth achieved approximately 80% confluency. hUC-MSCs at passage 3 were used for subsequent experiments. All experiments were approved by the Ethical Committee of the Affiliated Hospital of Qingdao University (approval No. QYFYWZLL 26435) on June 26, 2021. Informed consent was obtained from all donors prior to sample collection.

Adipogenesis, osteogenesis, and chondrogenesis of hUC-MSCs

Adipogenesis of hUC-MSCs was induced using the MesenCult™ adipogenic differentiation kit (STEMCELL Technologies, Seattle, WA, USA, Cat# 05412). hUC-MSCs at passage 3 were seeded at a density of 5 × 10³ cells/cm² in growth medium (containing DMEM, 10% fetal bovine serum, 1% penicillin-streptomycin, and 1% L-glutamine solution) and incubated at 37°C until they were approximately 90–100% confluent. The MesenCult™ adipogenic differentiation kit was then used, and the culture was changed two or three times per week for 21 days in total. Finally, cells were fixed and stained with fresh Oil Red O (KeyGEN BioTECH, Nanjing, China, Cat# KGA355) for 15 minutes to detect intracellular lipids.

To induce osteogenic differentiation, passage 3 hUC-MSCs were plated at 2 × 10⁴ cells/cm² and incubated overnight at 37°C with 5% CO₂. When cells achieved 60–70% confluence, the media was removed and replaced with 2 mL of reagent from the OriCell™ Osteogenesis Differentiation Kit (Cyagen, Guangzhou, China, Cat# HUXUC-90021) in each well. During 4 weeks of culture, the differentiation medium was regularly replaced. The cells were fixed and osteogenesis was evaluated using alizarin red S staining (Cyagen, Cat# HUXUC-90021), which was applied to assess the deposition of intracellular calcium.

To assess the capacity of hUC-MSCs to differentiate into chondrocytes, approximately 2 × 10⁶ passage 3 hUC-MSCs were centrifuged in a 15 mL polypropylene tube. The supernatant was removed and 0.5 mL fresh chondrogenesis induction differentiation medium was added to the pellet (STEMCELL Technologies, Cat# 05455). On day 2 and every 3 days afterward, the medium was carefully replaced with 0.5 mL of chondrogenesis induction differentiation medium without disturbing the pellet. After 4 weeks, the pellets were fixed, embedded in paraffin, and cut into 7-μm-thick sections. The sections were placed on slides, deparaffinized, rehydrated, and stained with Alcian Blue solution (KeyGEN BioTECH, Cat# KGA354).

Generation of NSCs from hUC-MSCs

Neurospheres are free-floating spherical aggregates that contain a mixture of NSCs/NPCs that retain their capacity to differentiate in neurons or glial cells under appropriate environmental conditions (Lindvall et al., 2004; Mukai et al., 2016). The generation of NSCs from hUC-MSCs was performed following previous reports with slight modifications (Mukai et al., 2016; Zhao et al., 2016; Peng et al., 2019). Briefly, hUC-MSCs were harvested as above and reseeded at a density of 1.5–2 × 10⁵ cells/cm² in low-attachment 6-well plates (Corning, Cat# 3471) with neurosphere-forming medium containing DMEM/nutrient mixture F-12 (Gibco, Cat# C11330500BT) supplemented with 20 ng/mL epithelial growth factor (PeproTech, Rocky Hill, NJ, USA, Cat# AF-100-15-100), 20 ng/mL basic fibroblast growth factor (PeproTech, Cat# AF-100-18B-50), and 2% B27 supplement (Gibco, 12587010). Cells were cultured at 37°C with 5% CO₂. As early as 4–6 hours after culture in neurosphere medium, the hUC-MSCs began to aggregate and gradually form sphere-like structures; the cells were frequently observed clumped together. Over several days of in vitro culture, almost all hUC-MSCs formed neurosphere-like structures, and the size of neurospheres increased from 3 to 7 days. The medium was changed and the diameters of neurospheres were measured on the 7th day under a light microscope with 100× magnification (Nikon, Tokyo, Japan, TE2000U) in six random visual fields. In some experiments, 5 and 10 nM NGR1β (R&D, Minneapolis, MN, USA, Cat# 396-HB) was added to the neurosphere medium during the induction process.

Immunofluorescence staining

The cells were fixed with 4% paraformaldehyde and washed with PBS. Cells were permeabilized with PBS with 0.5% Triton X-100 for 10 minutes and blocked with 10% goat serum (Boster, Wuhan, China, Cat# AR0009) for 40 minutes. The samples were then incubated with primary antibody (listed in Table 1) overnight at 4°C. The next day, the samples were incubated with secondary antibody (goat anti-rabbit Alexa Fluor 488 or anti-mouse Alexa Fluor 647, 1:200, Beyotime, Shanghai, China, Cat# A0423/A0473, RRID: AB_2891323/AB_2891322) at 37°C for 1 hour. Hoechst 33342 (Beyotime, Cat# C1025) was used to stain nuclei. Cells were visualized using a fluorescent microscope and the immunofluorescence intensity was measured by ImageJ software (version 1.52p; National Institutes of Health, Bethesda, MD, USA) (Schindelin et al., 2012).

Table 1.

Information of primary antibodies

Antibody	Supplier	Catalog No.	RRID No.	Species	Dilution	Applications
CD44	Abcam (Cambridge, UK)	ab189524	AB_2885107	Mouse, rat, human	1:200	Immunocytochemistry
					1:1000	Flow cytometry
CD90	Bioss (Beijing, China)	bs-0778R	AB_11097878	Human, mouse, rat	1:200	Immunocytochemistry
					1:1000	Flow cytometry
CD105	Abcam (Cambridge, UK)	ab231774	AB_2905493	Human	1:200	Immunocytochemistry
					1:1000	Flow cytometry
CD34	Abcam (Cambridge, UK)	ab81289	AB_1640331	Mouse, rat, human	1:1000	Flow cytometry
CD45	Abcam (Cambridge, UK)	ab40763	AB_726545	Human	1:1000	Flow cytometry
HLR-DA	Abcam (Cambridge, UK)	ab92511	AB_10563656	Human	1:1000	Flow cytometry
SOX2	Proteintech (Wuhan, Hubei Province, China)	11064-1-AP	AB_2195801	Human, mouse, rat, zebrafish	1:200	Immunocytochemistry
Nestin	STEMCELL (Seattle, WA, USA)	60091.1	AB_2905494	Human, non-human primate	1:200	Immunocytochemistry
β-ACTIN	Proteintech (Wuhan, Hubei Province, China)	20536-1-AP	AB_10700003	Human, mouse, rat, monkey, canine	1:1000	Western blot
GPX4	ZENBIO (Chengdu, Sichuan Province, China)	381958	AB_2905495	Human, mouse, rat,	1:500	Western blot
SLC7A11	ABclonal (Wuhan, Hubei Province, China)	A15604	AB_2763010	Rat	1:1000	Western blot
p53	ABclonal (Wuhan, Hubei Province, China)	A3185	AB_2764972	Human, mouse, rat	1:1000	Western blot
Nrf2	Affinity (Cincinnati, OH, USA)	AF7006	AB_2835314	Human, mouse, rat	1:500	Western blot

GPX4: Glutathione peroxidase 4; Nrf2: nuclear factor erythroid 2-related factor 2.

Flow cytometry analysis

When cells at passage 3 reached 80–90% confluency, the adherent cells were digested with 0.25% trypsin-ethylene diamine tetraacetic acid (Gibco, 25200056) and fixed in 80% ethyl alcohol. Cells were permeabilized with PBS with 0.5% Triton X-100 and then blocked. The cells were incubated with primary antibodies (listed in Table 1) overnight at 4°C, followed by incubation with goat anti-rabbit Alexa Fluor 488 (secondary antibody) for 60 minutes at 37°C on the next day. Finally, the samples were resuspended in 500 µL PBS and examined with a flow cytometer (Beckman, Coulter, Brea, CA, USA, CytoFLEX). Histograms were generated based on computed results using FlowJo V10 software (BD Biosciences, Franklin Lakes, NJ, USA).

5-Ethynyl-2’-deoxyuridine staining

Cell proliferation was evaluated by 5-ethynyl-2′-deoxyuridine (EdU) incorporation assays using a BeyoClick™ EdU-594 kit (Beyotime, Cat# C0078S) following the manufacturer’s instruction. Briefly, EdU at 10 μM was added to NSCs cultured in differentiation medium on the 6th day. The cells were fixed and permeabilized. The cells were stained with Click Additive Solution kit, and nuclei were stained by Hoechst 33342. Images were taken using a fluorescent microscope (Nikon, A1R), and the percentages of EdU-positive cells were analyzed by ImageJ software.

Quantitative polymerase chain reaction (qPCR)

Total RNA was extracted from cells using a Micro Sample RNA kit (SparkJade, Jinan, China, Cat# AC1001) following the manufacturer’s instructions. The HiScript II One Step qRT-PCR Probe kit (Vazyme, Nanjing, China, Cat# Q222-01) was used for PCR amplification and the QuantStudio™ 3 Real-Time PCR System (ThermoFisher, Shanghai, China, Cat# A28567) was used for PCR reactions. The amplification conditions were as follows: 50°C for 15 minutes, 95°C for 30 seconds, 40 cycles of 95°C for 10 seconds and 60°C for 30 seconds, and finally 95°C for 15 seconds, 60°C for 60 seconds and 95°C for 15 seconds. Three technical repetitions were performed for each PCR reaction and three biological repeats were performed for each result. The relative mRNA expression level of genes was calculated by the 2^–ΔΔCt method (Livak and Schmittgen, 2001; Rao et al., 2013), with gene expression normalized to β-actin mRNA and further normalized to expression in the control group. The specific primer sequences (forward and reverse) are listed in Table 2.

Table 2.

Primers used for quantitative polymerase chain reaction

Genes	Genbanks	Forward primer sequences (5’–3’)	Reverse primer sequences (5’–3’)	Product length (bp)
β-ACTIN	NM_001101.5	CAT CCG CAA AGA CCT GTA CG	CCT GCT TGC TGA TCC ACA TC	218
NANOG	JX105036.1	CAG CCA AAT TCT CCT GCC AG	CAC GTC TTC AGG TTG CAT GT	153
4-OCT	NM_001285986.2	GGT CCG AGT GTG GTT CTG TA	CGA GGA GTA CAG TGC AGT GA	190
SOX2	NM_003106.4	TGA TGG AGA CGG AGC TGA AG	GCT TGC TGA TCT CCG AGT TG	210
BMI1	NM_005180.9	TTG TTT GCC TAG CCC CAG TA	GAA GTT GCT GAT GAC CCA	165
NESTIN	NM_006617.2	TCT TTG CTC CCA GTC CTG AG	GGG CTC TGA TCT CTG CAT CT	187
NEUROD1	NM_002500.5	GAG ACG CAT GAA GGC TAA CG	CTG AAC GAA GGA GAC CAG GT	208
PAX6	AH014790.2	GGT GTC TTT GTC AAC GGG	TAC CAC CGA TTG CCC TG	189

NeuroD1: Neurogenic differentiation 1; Oct4: octamer-binding transcription factor 4.

OGD/R model

PC12 cells (Cat# iCell-r026) purchased from iCell (Shanghai, China) were cultured in modified RPMI medium (Hyclone, Cat# SH30809.01) containing 10% fetal equine serum (Solarbio, Beijing, China, Cat# S9050), 5% fetal bovine serum (PAN, Aidenbach, Germany, Cat# ST30-3302), and 1% penicillin-streptomycin. After 24 hours of culture, medium was replaced with glucose-free and serum-free RPMI (Procell, Wuhan, China, Cat# PM150122). The cells were placed in an anaerobic chamber containing 95% N₂ and 5% CO₂ and incubated at 37°C for various times. Fresh complete culture medium was added to cells for reoxygenation and cells were maintained under normal conditions for various times (Figure 1)

Figure 1.

Experimental design of the in vitro study.

NC: Negative control; NSCs: neural stem cells; NRG1β: neuregulin 1β; OGD/R: oxygen-glucose deprivation/reoxygenation.

Cell counting kit-8 assay

The viability of PC12 cells was evaluated using the cell counting kit-8 (Beyotime, Cat# C0038) assay. Briefly, PC12 cells (5 × 10⁴/mL) were plated in a 96-well plate with six replicates in each group. After stimulation through OGD/R, cell counting kit-8 solution was added to plates (10 μL/well). The optical density values were measured at 450 nm by an automatic microplate reader (BioTek, Winooski, VT, USA, Synergy™ H1).

ROS detection

An ROS assay kit (Beyotime, Cat# S0033S) was used to detect the level of intracellular ROS production. PC12 cells were incubated with 10 μM dichlorodihydrofluorescein diacetate probes for 30 minutes at 37°C following the manufacturer’s instructions. Fluorescence intensity was detected using a fluorescence microscope (Nikon), and the fluorescence signal of each group was quantified by ImageJ software.

Cell co-culture model

The co-culture system using Transwell chambers was conducted as previously reported (De Simone et al., 2017). Using a two-chamber Transwell system separated by a polycarbonate membrane with a 0.4 μm pore size (Corning, Cat# 3412/3413), neurospheres (approximately 1 × 10⁶ cells in one chamber) were cultured in the upper compartment and PC12 cells were grown in the bottom well of the chamber. After co-culture for 48 hours, cells were collected for further experiments.

Transmission electron microscopy

PC12 cells in different treatment groups were collected and immediately fixed in 0.2 M PBS (pH 7.2) containing 2.5% glutaraldehyde overnight at 4°C. Following standard procedures for transmission electron microscopy, the samples were processed by a series of alcohol washes for dehydration; infiltration was performed with propylene oxide and resin, followed by embedding. The samples were sectioned at a thickness of 50 nm with an ultramicrotome (Leica, Wetzlar, Germany, EMUC7). The sections were transferred to a copper grid, and the cellular organelles were stained with lead citrate and uranium. Images were obtained and mitochondrial morphology was examined using an HT7700 transmission electron microscope (Hitachi, Tokyo, Japan) with the setting of accelerating voltage at 80 kV.

Detection of malondialdehyde (MDA), superoxide dismutase (SOD), GSH/oxidized glutathione (GSSG) and NADPH/NADP⁺ levels

The levels of MDA, SOD, GSH/GSSG, and NADPH/NADP⁺ in PC12 cells were measured using MDA (Beyotime, Cat# S0131S), SOD (Solarbio, Cat# BC0175), GSH/GSSG (Beyotime, Cat# S0053), and NADPH/NADP⁺ (Nanjing Jiancheng, Nanjing, China, Cat# A115-1-1) assay kits, respectively, following the manufacturers’ protocols. The quantity of total protein was measured with a BCA Protein Assay Kit (CWBIO, Taizhou, Jiangsu, China, CW0014). Absorbance was measured by a spectrometry automatic microplate reader (BioTek).

Western blot assay

PC12 cells were lysed in radio immunoprecipitation assay lysis buffer (Beyotime, Cat# P0013C). Protein samples were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to a polyvinylidene fluoride membrane (Millipore, Temecula, CA, USA, Cat# ISEQ00010). The membranes were blocked and then incubated with primary antibodies (listed in Table 1) at 4°C overnight. The next day, the membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG (1:1000, Beyotime, Cat# A0208, RRID: AB_2892644) for 2 hours at approximately 25°C. The protein bands were visualized with the BeyoECL plus kit (Affinity, Cincinnati, OH, USA, KF001), and the relative protein expression (relative to β-actin) was analyzed by Image J software.

Small interfering RNA (siRNA) transfection

siRNA against SLC7A11 (siRNA-SLC7A11) and negative control (NC) siRNA were purchased from GenePharma Co., Ltd. (Shanghai, China). Transfection was performed using Lipofectamine 8000 (Beyotime, Cat# C0533) following the manufacturer’s instructions. Briefly, after PC12 cells were cultured for 24 hours, the medium was replaced with serum-free medium. siRNA-SLC7A11 and Lipofectamine 8000 was mixed 20 minutes before addition into the serum-free medium. After 48 hours of transfection, the knockdown efficiency was determined by western blotting; ACTIN was used for normalization.

Statistical analysis

The evaluators were blinded to the assignments. Data are shown as the mean ± standard deviation (SD). Statistical analysis of two groups was performed with SPSS 22.0 (IBM, Armonk, NY, USA), and significant differences were determined by one-way analysis of variance followed by Dunnett’s post hoc test. P < 0.05 was considered statistically significant.

Results

Culture and characterization of hUC-MSCs

UC tissues from healthy full-term pregnant woman were isolated and cultured, and primary hUC-MSCs were generated as described in Methods. As shown in Figure 2A, hUC-MSCs were arranged in parallel or a spiral shape. Cells were generally passed every 3 days because of their rapid proliferation. Immunofluorescence analysis demonstrated that hUC-MSCs were immunopositive for characteristic markers CD44, CD90, and CD105 (Figure 2B). To further characterize the hUC-MSCs, passage 3 cells were examined human MSC surface markers CD105, CD90, CD44, CD34, CD45 and HLA-DR, as defined by the International Therapy of Cellular Therapy, with flow cytometry. Over 90% of the cells exhibited positive surface expression for CD105, CD90 and CD44; fewer cells were positive for CD34, CD45 and HLA-DR (Figure 2C). These data show that the obtained hUC-MSCs exhibited similar morphology and surface marker expression compared with MSCs. To investigate the differentiation potential of hUC-MSCs, cells were induced into adipocytes, osteoblasts, and chondrocytes. An accumulation of Oil Red O-stained lipid drops was observed in differentiated hUC-MSCs, indicating adipogenesis; osteogenesis was observed with significant calcium deposition determined by alizarin S staining in treated cells, and chondrogenesis was revealed by proteoglycan staining with Alcian blue and was proved by the presence of extracellular matrix formation (Figure 2D). These results showed that the hUC-MSCs had differentiation potential.

Figure 2.

Characterization of hUC-MSCs.

(A) Morphology of passage 3 hUC-MSCs after 0, 24, 48, and 72 hours of culture. hUC-MSCs were spindle-like cells and uniformly distributed on the bottom of plate. Scale bars: 100 μm. (B) CD44, CD90, and CD105 (green) staining of hUC-MSCs. The nucleus was counterstained with Hoechst 33342 (blue). Scale bars: 50 μm. (C) Flow cytometry analysis of hUC-MSCs for MSC positive markers: CD105, CD90, and CD44 (top row), and MSC negative markers: CD34, CD45, and HLA-DR (bottom row). The histograms in the right panel show percentages of positive cells. Data are presented as mean ± SD and three biologically independent repeats for each group. (D) hUC-MSCs were successfully differentiated into adipocytes (arrow, left, stained with Oil Red O), osteoblasts (arrow, middle, stained with Alizarin Red staining), and chondrocytes (right, stained with toluidine blue staining). Scale bars: 100 μm. hUC-MSCs: Human umbilical cord mesenchymal stem cells; NC: negative control; P3: passage 3.

Effects of NRG1β on the generation of NSCs from hUC-MSCs

hUC-MSCs were induced into NSCs through the addition of growth factor, and NRG1β was added during induction in the pretreatment group, as described in Methods section. As shown in Figure 3A, neurospheres usually have a round shape, dense core, and clear outline; some cell clusters also show an irregular shape. On the 7^th day after induction of hUC-MSCs into NSCs, NSCs were collected and identified by assessing the key markers associated with pluripotency and neural progenitor cells. As shown in Figure 3B, qPCR showed that NSCs expressed the neural progenitor cell markers NESTIN, SOX2, PAX6, and NEUROD1 as well as pluripotency markers NANOG, OCT4, and BMI1. These genes were upregulated in the 5 and 10 nM NRG1β groups compared with the control group, with a larger difference in the 10 nM NRG1β group than that in the 5 nM NRG1β group (P < 0.05 or P < 0.01). Immunofluorescence analysis demonstrated that NSCs were also positive for SOX2 and Nestin, and the expressions of SOX2 and Nestin in the 10 nM NRG1β group were significantly higher than those in the 5 nM NRG1β and control groups (Figure 4A and B; P < 0.01 and P < 0.05). Together, these data suggested that NRG1β enhances the differentiation of hUC-MSCs into NSCs.

Figure 3.

NRG1β enhances the differentiation of hUC-MSCs to NSCs.

(A) hUC-MSCs aggregate to form neurospheres within 3 to 7 days in suspension. Scale bars: 100 μm. (B) qPCR analysis of the expression level of NSCs markers, NESTIN, NEUROD1, PAX6, and SOX2, and pluripotency markers: NANOG, OCT4, and BMIL1. Data were normalized to controls and are presented as mean ± SD. The experiment was repeated three times. *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Dunnett’s post hoc test). hUC-MSCs: Human umbilical cord mesenchymal stem cells; NeuroD1: neurogenic differentiation 1; NRG1β: neuregulin1β; NSCs: neural stem cells; Oct4: octamer-binding transcription factor 4; qPCR: quantitative polymerase chain reaction.

Figure 4.

NRG1β enhances expressions of key markers of NSCs.

(A) Nestin (red) and SOX2 (green) expression in the control and 5 and 10 nM NRG1β groups determined by immunostaining on the 7^th day after induction of hUC-MSCs into NSCs. All groups showed positive Nestin and SOX2 expression; the fluorescence intensity in the 10 nM NRG1β group was higher than that in the 5 nM NRG1β and control groups. Scale bars: 50 μm. (B) Immunopositive staining of SOX2 and Nestin. Five visual fields were randomly selected from each group; all experiments were repeated three times. Data are presented as the mean ± SD. *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Dunnett’s post hoc test). NRG1β: Neuregulin1β; NSCs: neural stem cells.

Statistical results showed that the diameter of neurospheres was significantly increased in the 5 and 10 nM NRG1β groups on the 7^th day compared with that in the control group (P < 0.05). The percentage of neurospheres over 80 μm in diameter in the 10 nM NRG1β group on the 7^th day was higher than those in the control and 5 nM NRG1β groups (Figure 5A). We further performed EdU assays and the results showed that the percentages of EdU-positive cells in the 5 and 10 nM NRG1β groups were significantly higher than that in the control group, with a significantly greater increase in the 10 nM NRG1β group than that in the 5 nM NRG1β group (P < 0.01; Figure 5B). Together, these findings confirmed that NRG1β positively regulates the proliferation of NSCs, and 10 nM NRG1β exhibited superior effects on the generation and proliferation of NSCs compared with 5 nM NRG1β.

Figure 5.

NRG1β enhances the proliferation potential of NSCs.

(A) Diameter and percentage of different neurospheres on the 7^th day in the control, 5 nM, and 10 nM NRG1β groups; each dot represents one biological replicate. (B) Representative images of NSCs proliferation activity determined using EdU assay on the 7^th day. Red indicates EdU-positive cells, and Hoechst 33342 (blue) was used for nuclei staining. The number of EdU-positive cells in the 5 and 10 nM NRG1β groups was higher than that in the control group. (C) Percentage of EdU-positive cells among total cells in different groups; the percentage of EdU-positive cells in the 5 and 10 nM NRG1β groups was higher than that in the control group. Data are expressed as mean ± SD. The experiment was repeated four times. *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Dunnett’s post hoc test). EdU: 5-Ethynyl-2′-deoxyuridine; NRG1β: neuregulin1β; NSCs: neural stem cells.

Pretreatment of NSCs with 10 nM NRG1β leads to a strong neuroprotective effect on damaged PC12 cells

We next used OGD/R as an in vitro model for I/R injury. Cells were cultured in anaerobic conditions for 2, 4, 6, 8, 10, and 12 hours, followed by reoxygenation under normal conditions for 24, 48, and 72 hours (Figure 6A). Cell viability assays showed that cell viability dramatically declined with increasing OGD time. With reoxygenation treatment, the quality and viability of PC12 cells were also reduced (Figure 6A and B). From these results, we chose OGD for 6 hours and reoxygenation for 48 hours for subsequent experiments.

Figure 6.

Effects of NSCs induced by 10 nM NRG1β on PC12 cell viability after OGD/R.

(A) Cell viability of PC12 cells exposed to OGD/R was evaluated by cell counting kit-8 assay. (B) The morphology of PC12 cells after 6 hours of OGD and reoxygenation for the indicated times. Compared with the control group, the OGD/R group had a poor adherent growth state, and the cells grew slower with reoxygenation time. Scale bars: 100 μm. (C) Morphology of PC12 cells after 6 hours of OGD and 48 hours of reoxygenation under the indicated co-culture condition. Network connections between cells formed in the NSCs and NSCs + 10 nM NRG1β group; dead cells were observed in the OGD/R and 10 nM NRG1β groups. Scale bars: 100 μm. (D) Quantification of PC12 cell viability after co-culture with NRG1β, NSCs and NSCs + 10 nM NRG1β shown in Figure 6C. The viability of PC12 cells was higher in NSCs and NSCs + 10 nM NRG1β groups than that in the OGD/R group. (E) Representative images of DCHF-DA fluorescence (green) in PC12 cells. Green fluorescence was strongest in the OGD/R group, followed by 10 nM NRG1β group, NSCs and NSCs + 10 nM NRG1β groups. Scale bars: 50 μm. (F) Fluorescence intensity of ROS in the NSCs + 10 nM NRG1β group was lower than that in the OGD/R group. Data are expressed as mean ± SD. All the experiments were repeated at least three times. *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Dunnett´s post hoc test). NRG1β: Neuregulin1β; NSCs: neural stem cells; OGD/R: oxygen-glucose deprivation/reoxygenation; ROS: reactive oxygen species.

To examine the effects of NSCs alone or with NRG1β on cell viability during OGD/R, the Transwell system was used to achieve co-culture of PC12 and NSCs. Co-culture condition was given immediately on reoxygenation and continuous for 48 hours; we found that PC12 cells formed an extensive connection in NSCs and NSCs + 10 nM NRG1β groups, but round shape cells and adhesion decreased in the OGD/R group (Figure 6C). The combination treatment led to an improvement in cell viability, and the difference in the viability of the NSCs + 10 nM NRG1β group compared with the OGD/R group was significant (Figure 6D).

Owing to cerebral I/R injury caused ROS generation and as ferroptosis is characterized by the accumulation of lipid ROS, we next examined the level of ROS. The data indicated that compared with ROS level in the OGD/R group, the production of ROS was decreased in the NSCs + 10 nM NRG1β group (P < 0.05; Figure 6E and F). No differences were observed with other groups.

Previous studies reported that lipid peroxidation and oxidative stress may play a central role in regulating the process of ferroptosis. Next, we measured the level of MDA (an important product of lipid peroxidation (Conrad et al., 2018; Piloni et al., 2021; Yuan et al., 2021)) and the expression of important intracellular antioxidants including GSH/GSSG, SOD, NADPH/NADP⁺, and Nrf2 in different groups. The results showed that OGD/R decreased the levels of GSH/GSSG (P < 0.05), SOD (P < 0.01), NADPH/NADP⁺ (P < 0.05), and Nrf2 (P < 0.01) in PC12 cells compared with results in the control group; co-culture with NSCs and NSCs + 10 nM NRG1β remarkably increased the contents of these indexes after OGD/R, with a more pronounced effect in the NSCs + 10 nM NRG1β group (P < 0.01; Figure 7A and B). MDA level was higher in the OGD/R group compared with the control group, but co-culture decreased the expression of MDA compared with expression in the OGD/R group (P < 0.05), and the difference in the NSCs + 10 nM NRG1β group was most significant among the 10 nM NRG1β, NSCs and NSCs + 10 nM NRG1β groups (P < 0.01; Figure 7A). Together, these data showed that pretreated NSCs with 10 nM NRG1β exhibit a strong neuroprotective effect on damaged PC12 cells. Morphologically, ferroptotic cells exhibit typical features of mitochondrial atrophy, diminished or vanished of mitochondrial cristae, and ruptured outer membrane. We also observed smaller mitochondria, increased membrane density and reduction/vanishing of mitochondria in damaged PC12 cells; mitochondrial damage improved in the co-culture groups (Figure 8A).

Figure 7.

NSCs pretreated with 10 nM NRG1β rescue PC12 cell damage during OGD/R.

(A) MDA, GSH/GSSG, SOD, and NADPH/NADP⁺ levels. (B) The expression level of Nrf2 was detected by western blotting. Data are expressed as mean ± SD. The experiments were repeated at least three times. *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Dunnett’s post hoc test). GSH/GSSG: Glutathione/oxidized glutathione; MDA: malondialdehyde; NADPH: nicotinamide adenine dinucleotide phosphate+; Nrf2: nuclear factor erythroid 2-related factor 2; NRG1β: neuregulin1β; NSCs: neural stem cells; OGD/R: oxygen-glucose deprivation/reoxygenation; SOD: superoxide dismutase.

Figure 8.

NRG1β relieves PC12 cell damage induced by OGD/R through regulating ferroptosis.

(A) TEM images. Red arrows indicate damaged mitochondria, and black arrows indicate normal mitochondria. Mitochondrial crest was broken and mitochondrial damage was aggravated, but this damage was improved after 10 nM NRG1β, NSCs, and NSCs + 10 nM NRG1β intervention. Scale bars: 2 μm (upper), 500 nm (lower). (B) The expression levels of p53, SLC7A11, and GPX4 protein were detected by western blotting, and the relative protein levels were normalized with ACTIN. Data are expressed as mean ± SD. The experiments were repeated three times. **P < 0.01 (one-way analysis of variance followed by Dunnett’s post hoc test). GPX4: Glutathione peroxidase 4; NRG1β: neuregulin1β; NSCs: neural stem cells; OGD/R: oxygen-glucose deprivation/reoxygenation; TEM: transmission electron microscopy.

Studies have shown that p53 suppresses the expression of SLC7A11 (which is the light chain subunit of System Xc^–). Reduced SLC7A11 leads to decreased synthesis of GSH and inactivation of GPX4, thus reduces cystine uptake and triggers ferroptosis (Liu et al., 2022). Western blot assay demonstrated that p53 level in the PC12 cells co-cultured with NSCs (OGD/R + NSCs) group and PC12 cells co-cultured with NSCs pretreated with 10 nM NRG1β (OGD/R + NSCs + 10 nM NRG1β) group was lower compared with levels in the OGD/R group (P < 0.01); no differences were observed between the OGD/R and OGD/R + 10 nM NRG1β groups. GPX4 and SLC7A11 levels were lower in the OGD/R group compared with the OGD/R + NSCs and OGD/R + NSCs + 10 nM NRG1β groups (P < 0.01; Figure 8B). These results suggested that the intervention group may play a protective role in reducing the level of ferroptosis in damaged PC12 cells through activating expression of p53, SCL7A11 and GPX4.

To further determine whether the protective role of NSCs pretreated with NRG1β on damaged PC12 cells occurs via the activating pathway of p53/SCL7A11/GPX4, SLC7A11 was knocked down using siRNA-SLC7A11 (Figure 9A and B). Western blotting confirmed a significant decrease in SLC7A11 expression after SLC7A11 knockdown (P < 0.05). We found that the expression levels of p53, SLC7A11 and GPX4 in the NSCs + 10 nM NRG1β group were significantly reversed compared with the OGD/R + siRNA-SLC7A11 group (P < 0.01; Figure 9C). These findings suggested that NSCs + 10 nM NRG1β play an intervention role in reducing levels of ferroptosis in damaged PC12 cells through activating the p53/SCL7A11/GPX4 pathway.

Figure 9.

Effects of NSCs + 10 nM NRG1β on ferroptosis caused by OGD/R in the presence of siRNA-SLC7A11.

(A) Fluorescence expression of negative control-FAM (NC-FAM) after transfection for 8 hours. (B) Western blotting was performed to detect protein expression in control, NC, and positive control (ACTIN) groups after transfection for 24 hours. (C) The expression of p53, SLC7A11 and GPX4 were detected by western blotting, and relative protein levels were normalized with ACTIN. Data are expressed as the mean ± SD. The experiments were repeated three times. *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Dunnett´s post hoc test). GPX4: Glutathione peroxidase 4; NC: negative control; NRG1β: neuregulin1β; NSCs: neural stem cells; OGD/R: oxygen-glucose deprivation/reoxygenation.

Discussion

Several therapies for cerebral I/R injury are available, including drugs and surgery, but these treatments have their limitations (Zhu et al., 2012). Cell therapy has become an emerging alternative treatment for cerebral I/R injury, especially therapies using stem cells (Lindvall and Kokaia, 2011; Scheibe et al., 2012; Liu et al., 2014b). Among various stem cell sources, NSCs are considered as an ideal candidate seed cell for stem cell-based treatment for cerebral I/R injury because they retain the ability to self-renew and can differentiate into the major cell types of the brain (Kelly et al., 2004; Boese et al., 2018). Nevertheless, technical difficulties associated with the isolation and propagation of NSCs and low rates of survival and neuronal differentiation limit their clinical application. It is therefore essential to find an alternative source of NSCs and strategies to increase the capacity of NSCs to adapt their fate and function in a changing pathological environment.

The issue of direct conversion of hUC-MSCs into NSCs/NPCs under certain conditions has not been addressed (Fu et al., 2008; Liu et al., 2014a; Mukai et al., 2016) Identifying a strategy to differentiate MSCs into NSCs will allow the generation of NSCs in vitro without being restricted by ethical issues is essential. At present, the survival and differentiation of NSCs are key issues affecting the efficacy of exogenous NSCs in the treatment of reperfusion injury. Therefore, many studies have attempted to use different strategies to influence these functions by expressing specific genes via viral transfection, pretreating NSCs with inflammatory immune factors, and combining NSCs with cytokines to increase the therapeutic effect of transplanted cells (Zhang et al., 2019). These methods provide a basis for improving the effect of exogenous NSC therapy.

In this study, we isolated hUC-MSCs and cultured cells with serum-free medium supplemented with epidermal growth factor and basic fibroblast growth factor. Cells showed formation of spheroid structures, and sphere colonies generated large neurospheres. Our research indicated that NRG1β enhanced the differentiation of hUC-MSCs to NSCs, and the effect of 10 nM NRG1β was more effective than that of 5 nM NRG1β. The proliferation of NSCs was increased in the presence of 5 nM and 10 nM NRG1β. However, the mechanism of how NRG1β promotes the differentiation and proliferation of NSCs remains elusive. We hypothesized that increasing the proliferation capacity of NSCs led to the increase in the number of NSCs, and thus the differentiation efficiency of hUC-MSCs to NSCs was also improved. This finding provides a new strategy for using in vitro interventions to improve the therapeutic effects of NSCs after transplantation.

We further assessed the neuroprotective effects of NSCs pretreated with NRG1β on damaged PC12 cells induced by OGD/R. Our results showed that the levels of ROS, MDA, GSH/GSSG, SOD, NADPH/NADP⁺ and Nrf2 and the level of mitochondrial damage were all improved in damaged PC12 cells after intervention with NSCs pretreated with NRG1β. The neuroprotective effects of NSCs and NSCs pretreated with 10 nM NRG1β were stronger than those in NRG1β groups. Importantly, the intervention effect in NSCs pretreated with 10 nM NRG1β, in terms of ROS, MAD and Nrf2, was better than in the NSCs group.

Transplantation of NSCs has been shown to regulate the immune response through a paracrine regulation response; NSCs release neurotrophic factor, vascular endothelial growth factor and other specific factors, which inhibit or reduce the inflammatory response after stroke. Endogenous NSCs regulate the local inflammatory microenvironment after transplantation and promote angiogenesis and synaptic plasticity by secreting nerve growth factor 4, laminin, integrins, and thrombospondins (Lladó et al., 2004; Staquicini et al., 2009; Horie et al., 2011). We speculate that 10 nM NRG1β may enhance the potential of NSCs and make them more resilient to impaired microenvironments through secreting some molecules to better respond to the pathological environment.

Studies have indicated that ferroptosis mediates ischemic injury of neurons by regulating the p53/SLC7A11/GPX4 pathway (Lan et al., 2020; Lu et al., 2020). As a subunit of System Xc^–, SLC7A11 plays a key role in the conversion of extracellular cysteine and intracellular glutamate, and the imbalance of glutamate further induces the inactivation of GPX4 (Cao and Dixon, 2016; He et al., 2021). GPX4 plays a pivotal role in the inhibition of lipid peroxidation and ROS recruitment, which ultimately leads to the occurrence of ferroptosis (Maiorino et al., 2018). SLC7A11 is a target of p53, which binds to p53-responsive element in the promoter region of SLC7A11 and represses its expression, further decreasing the uptake of extracellular cysteine and the synthesis of intracellular GSH and activating ferroptosis (Murphy, 2016; Liu et al., 2020). Our results showed that while the expressions of SLC7A11 and GPX4 were decreased in response to OGD/R, the expressions were increased in the 10 nM NRG1β, NSCs, and NSCs + 10 nM NRG1β groups, especially in the NSCs + 10 nM NRG1β group. In addition, the activation of p53 was inhibited in co-culture groups compared with that in the OGD/R group. siRNA-mediated knockdown of SLC7A11 showed that the NSCs + 10 nM NRG1β intervention attenuated ferroptosis induced by OGD/R in PC12 cells. These results provide evidence of the neuroprotective effects of NSCs pretreated with NRG1β during OGD/R and indicated that ferroptosis may be mediated by the activation of p53/SLC7A11/GPX4 in the process.

In summary, our data indicate that pretreatment of NSCs with 10 nM NRG1β enhanced their neuroprotective effects during OGD/R. In addition, our results suggest that ferroptosis is involved in the pathogenesis of OGD/R, and the activation of p53/SLC7A11/GPX4 pathway may decrease the level of ferroptosis in damaged PC12 cells. These findings may provide important evidence showing that NSCs obtained better therapy for cerebral I/R injury. However, the precise effects of NRG1β on the differentiation of NSCs derived from hUC-MSCs remain unclear, and whether the inner character of NSCs was affected by NRG1β should be examined. Additionally, further studies should investigate the effects of transplantation of stem cells pretreated with NRG1β into a cerebral I/R injury animal model to verify the neuroprotective effects and the related molecular mechanism.