This article is mentioned in:

Introduction

Neonatal hypoxic-ischemic brain damage (HIBD) is a common clinical syndrome in newborns caused by anoxia and reduced cerebral blood flow or temporary severance during the perinatal period (1–3). It has been reported that 2–6/1,000 newborns experience HIBD (4), and ~25% of infants who have HIBD develop neurologic sequela (5,6). Hypothermia is the only admitted therapy for treatment of neonatal HIBD in the clinic; however, the therapeutic window of hypothermia is confined to 6 h after birth, and even then, >40% of the infants either died or survive with various impairments following hypothermic treatment (7–9). Thus, alternative effective and safe therapies for treatment of HIBD are required.

Globins are hemoproteins that bind O2 and serve an important role in the animal's respiration and oxidative energy production (10,11). However, globins may also possess other functions, such as the decomposition of nitric oxide (NO), the detoxification of reactive oxygen species (ROS) or intracellular signaling (12). Cytoglobin (CYGB), the fourth member of the vertebrate globin family of hemoproteins, is ubiquitously expressed in various tissues and organs, including the liver, kidney, brain and retina (13). CYGB may serve a cytoprotective role under hypoxic and/or ischemic conditions (14–17). Previous studies have also reported that pretreatment of CYGB-overexpression reduces HI injury and improves long term memory and athletic ability following neonatal HIBD (18). However, a safe and effective method of upregulating CYGB expression in animals remains a challenge that requires further investigation.

Previous studies have revealed that several types of cells may serve important roles in relieving HIBD (19–21). Preclinical trials on stem cells in cerebral palsy have been conducted and have reported significant improvements in acute hypoxic injury animal models (22). Mesenchymal stem cells (MSCs) are well-known for their ‘immunosuppressive’ properties, and thus may be important candidates for allogeneic cell therapy (23). In recent years, human umbilical cord-derived MSCs (HuMSCs) have become an alternate source of MSCs. Moreover, studies have observed that transplantation of HuMSCs at an early stage following HIBD can reduce HI injury in rats and decrease gliosis (24,25). Stem cells can also act as a gene transporter in gene therapy (26). The use of stem cells as transgenic strategies does not require integration of the therapeutic DNA into the chromosomes of the patient's cells, and instead, the transferred DNA is stabilized extrachromosomally (27). However, it has been identified that there are major risks associated with the use of integrating vectors, including retroviral vectors, arising from their potential for insertional mutagenesis, in which the vector inserts into the DNA of a cell and disrupts a functional element of that DNA (27); it has been reported that stem cell transgenic therapy can reduce this risk (28).

The aim of the present study was to investigate whether nasal transplantation of CYGB genetically modified HuMSCs (CYGB-HuMSCs) exhibited higher protective effects in neonatal rats from HIBD compared with rats treated without CYGB genetically modified HuMSCs. Additionally, the potential underlying mechanism was examined.

Materials and methods

Preparation of HuMSCs

Ethical approval was obtained from the Institutional Review Board of Maternal and Child Health Care Hospital of Shenzhen University (approval no. SZPSFY2017-06). HuMSCs were prepared as previously described (29). With the written consent of the patients, human umbilical cords from 5 patients (age range, 23–34 years old) who underwent full-term caesarian sections in the Maternal and Child Health Care Hospital (Pingshan, Shenzhen, China) from February 2018 to March 2018 were collected immediately into sterilized 50 ml tubes. After washing with PBS for 30 sec, the human umbilical cord was cut into 2–3 cm thick sections. The umbilical arteries and veins were removed, and the remaining tissue, the Wharton's jelly, was dissected into smaller fragments and transferred to 75 cm2 flasks containing DMEM/F12 media (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (Gibco; Thermo Fisher Scientific, Inc.), 100 µg/ml penicillin/streptomycin (Sangon Biotech Co., Ltd.), 1 g/ml amphotericin B (Gilead Sciences, Inc.), 5 ng/ml epidermal growth factor (PeproTech, Inc.) and 5 ng/ml basic fibroblast growth factor (PeproTech, Inc.). Cultures remained undisturbed for 5–7 days at 37°C with 95% air/5% CO2 in a humidified incubator to allow migration of cells from the explants. Subsequently, the media was replaced. After three passages, cells were harvested. Cells were observed under a Zeiss Axio Imager Z1 light inverted microscope (magnification, ×100; Carl Zeiss AG).

Adenovirus-mediated transfection

Adenovirus plasmid carrying CYGB (Ad-CYGB) was purchased from Guangzhou Forevergen Technology Co., Ltd. Adenovirus plasmid carrying only enhanced green fluorescent protein (eGFP) was used as the transfection control. The transfection was performed using Lipofectamine® 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.). HuMSCs were seeded into 12-well plates at a concentration of 2×105 cells per well. After culture for 12 h, Ad-CYGB (2 µl; multiplicity of infection, 10) were added to the respective wells; the same volume of control adenovirus was added. On the second day after infection, the virus-containing medium was aspirated and replaced with fresh complete medium to continue the culture. The culture medium was replaced every 2 days thereafter for a total period of 7 days. Follow-up experiments were performed 48 h after the transfection. Cells were observed under a Zeiss Axio Imager Z1 fluorescent inverted microscope (magnification, ×200; Carl Zeiss AG).

Experimental animals

Sprague-Dawley rats (age, 7 days; 64 females; 56 males; weight, 12.5-16.3 g) were obtained from the Experimental Animal Center of Shantou University Medical College. The present study was approved by the Institutional Animal Care and Use Committee of Shenzhen University (approval no. 2017-06) and strictly adhered to the ARRIVE guidelines (30). All animals were maintained in a temperature- and humidity-controlled room (temperature, 22–24°C; humidity, 40–70%) with a 12 h light/dark cycle, and free access to food and water. Throughout the experiment, all animals were treated in accordance with the Guide for the Care and Use of Laboratory Animals in Shenzhen University, and all procedures conformed to internationally accredited guidelines and ethical regulations on animal research. Surgery was performed under isoflurane anesthesia, and all efforts were made to reduce the total number of animals used and minimize their potential suffering.

Animal groups

A total of 120 Sprague-Dawley rats were used in the present experiment. The rats were divided into four groups, with 30 rats in each group. Sham group animals received only anesthesia and exposure of the left common carotid artery, but no ligation and hypoxia operation. Animals in the HIBD, HuMSCs and CYGB-HuMSCs groups received ligation of the left common carotid artery combined with hypoxia. A total of 30 min after HIBD model preparation, animals in the HuMSCs group and CYGB-HuMSCs group received intranasal administration of 1×106 HuMSCs or CYGB-HuMSCs (3 µl), respectively. Animals in the HIBD group received only PBS (3 µl) via intranasal administration.

Establishment of an animal model of HIBD and transplantation of HuMSCs

The HIBD neonatal rat model was based on the classical Rice-Vannucci model (31). The experimental rat was deeply anesthetized via inhalation of isoflurane (3% to effect). Then, the left common carotid artery was exposed, followed by double-ligation with 5-0 silk sutures and the artery between the ligations was cut off. The wound was then sutured. The total time for surgery in each animal was ≤3 min. The body temperature of the animal was maintained at 37°C with a radiant warmer table. After surgery, animals were allowed 1–2 h to recover from anesthesia with their mother. After recovery, animals were placed in a container with a lowered oxygen percentage (8% oxygen balanced with 92% nitrogen) and placed in a 37°C water bath for 2.5 h to induce systemic hypoxia. Animals in the sham group only received anesthesia and exposure of the left common carotid artery. A total of 30 min after HI, 1×106 cells HuMSCs or CYGB-HuMSCs (3 µl) were intranasally administered, respectively. As described previously (32), prior to the administration of HuMSCs, to increase the permeability of the nasal mucosa, nostrils were treated with 3 µl hyaluronidase (100 U; Sigma-Aldrich; Merck KGaA) in PBS. A total of 30 min later, animals were administered 3 µl HuMSCs, CYGB-HuMSCs (1×106 cells) or PBS (3 µl) twice in each nostril.

Observation

After modeling for 3, 14 and 29 days, five randomly selected rats were sacrificed via inhalation anesthesia with isoflurane (3% to effect) followed by decapitation in each group. Then, reverse transcription-quantitative (RT-q)PCR and western blotting were used to detect CYGB expression in brain tissues. Additionally, ERK, phosphorylated-(p-)ERK, JNK, p-JNK, p38 and p-p38 protein expression levels were assessed using western blotting. Nissl staining results of the cortex and hippocampal Cornu Ammonis 1 (CA1) area of rats in each group were compared after modeling for 3 days. TUNEL assays and immunofluorescence were performed on day 3 after modeling; immunofluorescence was also performed on day 14. Moreover, a Morris-Water maze experiment was used 29 days after modeling.

Nissl staining

The obtained brain specimens were fixed in 4% (w/v) formaldehyde at room temperature for 36 h, paraffin-embedded and sliced into 3-µm thick sections. The paraffin sections were deparaffinized in xylene and rehydrated in a descending ethanol series at room temperature, then, the sections were immersed in 1% toluidine blue stain at 37°C. After 5 min, the sections were washed with water, differentiated with 1% glacial acetic acid and placed in xylene for 5 min at room temperature. Then the slices were removed from the xylene, dried and sealed with neutral gum. Samples were observed under a light microscope (Olympus BX43; Olympus Corporation) at ×200 magnification.

TUNEL staining

TUNEL staining was performed on paraffin sections using an in-situ cell death detection kit (DeadEnd™ Fluorometric TUNEL system; Promega Corporation), according to the manufacturer's protocol. Paraffin sections were prepared and fixed as previously described for Nissl staining. Paraffin sections were dewaxed in xylene and rehydrated in a descending ethanol series at room temperature, then treated with 20 mg/l DNase proteinase K at 20–37°C for 15 min. Subsequently, the sections were washed with PBS three times. Sections were then covered with 100 µl equilibration buffer and incubated at room temperature for 5–10 min. After removing the balance solution, 50 µl rTdT incubation buffer was added and the sections covered with plastic coverslips to incubate at 37°C for 60 min. Sections were then immersed in 2X saline sodium citrate in a dyeing tank at room temperature for 15 min. To stop the reactions, sections were washed twice with PBS. Subsequently, sections were stained with 1X Hoechst nuclear stain at 37°C for 15 min and mounted with neutral gum. Samples were observed under a Zeiss Axio Imager Z1 fluorescent microscope (magnification, ×200; Carl Zeiss AG) in three randomly selected fields of view. Image-Pro-Plus software (version 6.0; Media Cybernetics, Inc.) was used to perform the semi-quantitative analysis of the apoptotic cells.

Immunofluorescence

Paraffin sections were prepared and fixed as previously described for Nissl staining. The paraffin sections were deparaffinized in xylene and rehydrated using a descending ethanol series at room temperature. Then, the sections were permeabilized with 0.1% Triton X-100 at room temperature for 10 min. Sections were blocked with normal goat blocking serum (Abbkine Scientific Co., Ltd.) at a volume fraction of 3% at room temperature for 1 h. After washing three times with PBS for 2 min/time, sections were treated overnight with primary antibody at 4°C. The primary antibodies used were glial fibrillary acidic protein (GFAP; Cell Signaling Technology, Inc.; cat. no. 80788S; 1:50) and neuron-specific enolase (NSE; ProteinTech Group, Inc.; cat. no. 10149-1-AP; 1:50). The sections were rinsed twice with PBS and then incubated with the secondary Alexa Fluor 555 antibody (Abcam; cat. no. ab150074; 1:200) for 1 h at room temperature in the dark. Samples were observed under a Zeiss Axio Imager Z1 fluorescent microscope (magnification, ×200; Carl Zeiss AG).

RT-qPCR

After HI modeling for 3, 14 or 29 days, rats from different groups were anesthetized and decapitated. Samples of injured brain tissues were collected rapidly and total RNA was prepared from samples collected using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) as previously described (33). Total RNA (2 µg) was reverse transcribed using M-MLV reverse transcriptase (Promega Corporation). In a sterile RNase-free microcentrifuge tube, 2 µg total RNA and 1 µg primer in a total volume of 15 µl in water were added. The tube was heated tube to 70°C for 5 min. Then, the tube was cooled immediately on ice and centrifuged (500 × g; 4°C; 1 min) briefly to collect the solution at the bottom of the tube. In total, 5 µl 5X buffer, 6.25 µl 2 mM dNTP, 1 µl M-MLV and 12.75 µl RNase-free water were added to the tube, which was mix gently and incubated for 60 min at 42°C. cDNA expression levels were determined via qPCR using specific primers and GoTaq® qPCR Master mix (Promega Corporation) in a StepOne Plus amplifier (Bio-Rad Laboratories, Inc.). The thermocycling protocol was as follows: Initial denaturation at 95°C for 120 sec; followed by 40 cycles at 95°C for 15 sec, 60°C for 30 sec and 72°C for 45 sec, and then a final elongation at 72°C for 10 min. Data were quantified using the 2−ΔΔCq method (34). The primers used were as follows: CYGB (NM_134268.5; product length, 152 bp) forward, 5′-TTGCCAGTGACTTCCCACC-3′ and reverse 5′-CCCGAAGAGGGCAGTGTG-3′; and GAPDH (NM_001289745.3; product length, 185 bp) forward, 5′-GAGTCAACGGATTTGGTCG-3′ and reverse 5′-GAGTCAACGGATTTGGTCGT-3′.

Western blotting

Brain tissues collected simultaneously for the qPCR experiments were homogenized in ice-cold RIPA lysis buffer (Beyotime Institute of Biotechnology), and the protein concentrations were measured using a bicinchoninic acid protein assay kit (Beyotime Institute of Biotechnology). A total of 40 µg total protein per lane was loaded on a 12% SDS-gel, resolved using SDS-PAGE and transferred to a PVDF membrane (EMD Millipore). The membrane was blocked with TBS-Tween (20 mM Tris; pH 7.6; 135 mM NaCl; 0.05% Tween) containing 5% non-fat dry milk overnight at 4°C, washed with 1X TBS-0.1% Tween-20 and then incubated at 4°C with primary antibodies in blocking solution. The following primary antibodies were used: GAPDH (1:10,000; cat. no. 10494-1-AP; ProteinTech Group, Inc.), CYGB (1:1,000; cat. no. 13317-1-AP; ProteinTech Group, Inc.), p-p38 (1:1,000; cat. no. 4511S; Cell Signaling Technology, Inc.), p38 (1:1,000; cat. no. 8690S; Cell Signaling Technology, Inc.), p-ERK (1:1,000; cat. no. 4370; Cell Signaling Technology, Inc.), ERK (1:1,000; cat. no. 9102; Cell Signaling Technology, Inc.), p-JNK (1:1,000; cat. no. 4668; Cell Signaling Technology, Inc.) and JNK (1:1,000; cat. no. 9252; Cell Signaling Technology, Inc.). After washing the membrane four times with TBS-Tween, the horseradish peroxidase-labeled secondary goat anti-rabbit IgG antibody (1:8,000; cat. no. RS0002; Immunoway; Suzhou Ruiying Biotechnology Co., Ltd.), was added and incubated for 2 h at room temperature, and then washed again. Protein bands were visualized using an ECL Plus chemiluminescence kit (Guangzhou Forevergen Technology Co., Ltd.). ImageJ software version 1.52 h (National Institutes of Health) was used to perform the densitometric analysis.

Morris-Water maze test

The water maze was composed of a cylindrical pool and a platform. The height of the pool was 70 cm and the diameter was 80 cm. The diameter of the platform was 8 cm. Over the pool, there was a digital camera connected to the computer. The pool was filled with water until the surface of the water was ~0.5 cm above the surface of the platform, the water temperature was controlled at 22.0±0.5°C. A specific point on the pool was used as the point of entry for rats. The platform was placed in the third quadrant, and the position of the platform was unchanged throughout the entire experimental process. Training twice a day started 4 days prior to the formal experiment; rats were placed in the water from the four quadrants. If the animal found the platform within 120 sec, it was left on the platform for 20 sec. If the animal did not find the platform, it was placed on the platform and left for 20 sec. On the second day after the training finished, experiments were performed to record the time and the path length (PL) from the point of entry to the platform in 120 sec.

Statistical analysis

Statistical analysis was performed using GraphPad Prism version 8 (GraphPad Software, Inc.). Continuous experimental data are presented as the mean ± standard deviation of ≥3 experimental repeats. A one-way ANOVA followed by Tukey's of Bonferroni post hoc test was used to compare the differences between multiple groups. P<0.05 was considered to indicate a statistically significant difference.

Results

Morphology of cultured HuMSCs and transfection

After 5–7 days of culture, fibroblast-like cells migrated out from the surrounding tissues. Primary HuMSCs were passaged when they were cultured for 10–14 days and the cells reached 80% confluence. In the 3rd generation, cells had a stable fibroblast-like morphology (Fig. 1A). The results of eGFP immunofluorescence analysis demonstrated that the adenovirus plasmid transfection efficiency was ~80% (Fig. 1B and C). Furthermore, RT-qPCR results suggested that the transfected CYGB gene was efficiently expressed in HuMSCs (Fig. 1D and E).

Figure 1. Morphology of cultured HuMSCs. (A) HuMSCs after the third passage exhibited a fibroblast-like morphology. Magnification, ×100. eGFP fluorescence of HuMSCs transfected with (B) Ad-eGFP or (C) Ad-CYGB. Magnification, ×200. (D) mRNA and (E) protein expression levels of CYGB in HuMSCs transfected with Ad-eGFP or Ad-CYGB after 48 h. ***P<0.001 vs. HuMSC. HuMSC, human umbilical cord-derived mesenchymal stem cells; eGFP, enhanced green fluorescent protein; Ad-eGFP, Adenovirus plasmid carrying eGFP; CYGB, Cytoglobin.

Observation of histological changes in the brain using Nissl staining after modeling for 3 days

Neuronal cell loss in both the cerebral cortex and hippocampus was observed via Nissl staining after modeling for 3 days. HI markedly reduced the number of cells and resulted in irregularly arranged and smaller neurons in the brain 3 days after HI compared with the sham group (Fig. 2). However, in the HuMSCs group, nerve cells were regularly arranged and survival numbers were notably increased compared with the HI group (Fig. 2). Furthermore, high numbers of Nissl-stained cells were observed in the CYGB-HuMSCs group compared with the HuMSCs group (Fig. 2).

Figure 2. Morphology of neuronal cells in both the cortex and hippocampus CA1 area 3 days after HI and in the sham group. Magnification, ×200. HI, hypoxia-ischemia injury; CA, Cornu Ammonis. HIBD, hypoxic-ischemia brain damage; HuMSC, human umbilical cord-derived mesenchymal stem cells; CYGB, Cytoglobin.

TUNEL staining

To determine whether HuMSCs and CYGB-HuMSCs transplantation prevented apoptosis at the acute stage of neonatal HI injury, TUNEL staining of tissue sections was performed after 3 days of modelling (Fig. 3A and B). Significantly fewer apoptotic cells were observed in the HuMSCs group compared with the HI groups (P<0.01), and the number of apoptotic cells was significantly lower in the CYGB-HuMSCs group compared with the HuMSCs groups (Fig. 3A-C).

Figure 3. Levels of neural cell apoptosis 3 days after HI. (A) TUNEL staining was performed to evaluate neural cell apoptosis at an acute stage of neonatal HI injury and in the sham group (n=5). Magnification, ×100. (B) TUNEL staining on day 3 after HI and in the sham group. Magnification, ×200. (C) Significantly fewer apoptotic cells were observed in the HuMSCs group compared with the HIBD group. The number of apoptotic cells was significantly lower in the CYGB-HuMSCs group compared with the HuMSCs groups. aP<0.01 vs. Sham group at the same time; bP<0.01 vs. HIBD group at the same time; cP<0.01 vs. HuMSCs group at the same time. HI, hypoxia-ischemia injury; HuMSC, human umbilical cord-derived mesenchymal stem cells; CYGB, Cytoglobin; HIBD, hypoxic-ischemia brain damage.

Immunofluorescence

To assess whether HuMSCs or CYGB-HuMSCs underwent cell replacement via differentiation into neuron-like cells, immunofluorescence co-localization analysis was performed 3 and 14 days after HI. It was identified that HuMSCs or CYGB-HuMSCs were present and scattered in the damaged brain tissue (Fig. 4). Moreover, no eGFP and GFAP or EGFP and NSE double-stained positive cells were observed and no substantial differences were observed between 3 and 14 days.

Figure 4. Distribution, survival and differentiation of HuMSCs or CYGB-HuMSCs 3 and 14 days after HI. HuMSCs or CYGB-HuMSCs were scattered in the damaged brain tissue. No eGFP and GFAP or eGFP and NSE double stained positive cells were observed. Magnification, ×100. HuMSC, human umbilical cord-derived mesenchymal stem cells; HI, hypoxia-ischemia injury; CYGB, Cytoglobin; eGFP, enhanced GFP; GFAP, glial fibrillary acidic protein; NSE, neuron-specific enolase.

Expression of CYGB after transplantation of HuMSCs

To investigative the mRNA and protein expression levels of CYGB following CYGB-HuMSCs transplantation, RT-qPCR and western blotting were performed on samples collected from all groups following HI modeling for 3, 14 and 29 days. RT-qPCR results demonstrated that CYGB mRNA expression was significantly increased in the CYGB-HuMSCs group compared with the HIBD and HuMSCs groups (Fig. 5C) at all experimental time points. CYGB mRNA expression was significantly increased in the HuMSCs group compared with the HIBD group, but significantly decreased compared with the CYGB-HuMSC group at 14 and 29 days (Fig. 5C). Western blotting results identified that CYGB protein expression was also increased in the CYGB-HuMSCs group compared with the HIBD and HuMSCs groups. At all experimental time points, CYGB protein expression was also increased in the HuMSCs group compared with the HIBD group, while decreased compared with the CYGB-HuMSCs group. The expression levels of CYGB protein in both the HuMSCs and CYGB-HuMSCs groups gradually declined from 3 to 29 days (Fig. 5A and B).

Figure 5. Expression of CYGB in all groups after HI modeling for 3, 14 and 29 days (n=5 per group). (A) Expression of CYGB protein was assessed using (B) western blotting. GAPDH was used as the internal control. (C) mRNA expression of CYGB mRNA was determined using reverse transcription-quantitative PCR. The results demonstrated that both the protein and mRNA expression levels of CYGB were increased in the CYGB-HuMSCs group compared with the HIBD and HuMSCs groups. aP<0.01 vs. Sham group at the same time; bP<0.01 vs. HIBD group at the same time; cP<0.01 vs. HuMSCs group at the same time; a′P<0.05 vs. Sham group at the same time; b′P<0.05 vs. HIBD group at the same time. HI, hypoxia-ischemia injury; CYGB, Cytoglobin; HuMSC, human umbilical cord-derived mesenchymal stem cells; HIBD, hypoxic-ischemia brain damage.

Morris-Water maze test

The Morris-Water maze experiment was performed 29 days after HI to evaluate long term learning and memory, which is dependent on the function of the hippocampus and cortex (18). The results indicated a longer escape latency (EL) in the HIBD group compared with the sham group. EL in the HuMSCs group was significantly shorter compared with the HIBD group, and in the CYGB-HuMSCs group, EL was shorter compared with the HuMSCs group (Fig. 6A). The PL was also recorded. PL was longer in the HIBD group compared with the sham group (Fig. 6B). Furthermore, HuMSCs transplantation significantly reduced PL compared with the HIBD rats, and CYGB-HuMSCs transplantation resulted in further reduced PL values compared with HuMSCs transplantation. These results indicated a long-term neuroprotective effect of either HuMSCs or CYGB-HuMSCs on HIBD.

Figure 6. Morris-Water maze experiments 29 days after HI (n=5 per group). (A) EL values of the HI group were longer compared with the sham group. EL values in the HuMSCs group was significantly shorter compared with the HI group, and the EL values in the CYGB-HuMSCs group were shorter compared with the HuMSCs group. (B) PL values were longer in the HI compared with the sham group. In the HuMSCs group, PL was shorter compared with the HI group, and the PL value in the CYGB-HuMSCs group was also shorter compared with the HuMSCs group. aP<0.01 vs. Sham group at the same time; bP<0.01 vs. HIBD group at the same time; cP<0.01 vs. HuMSCs group at the same time; b′P<0.05 vs. HIBD group at the same time; c′P<0.05 vs. HuMSCs group at the same time. HI, hypoxia-ischemia injury; CYGB, Cytoglobin; HuMSC, human umbilical cord-derived mesenchymal stem cells; EL, escape latency; PL, path length; HIBD, hypoxic-ischemia brain damage.

CYGB-HuMSCs treatment modulates the p38 pathway

As demonstrated by the western blotting results (Fig. 7), compared with the Sham group, the ratio of p-p38/p38 was significantly increased in the HIBD group after HI modeling for 3, 14 and 29 days (Fig. 7A-C), suggesting that hypoxia and oxidative stress mediated the stimulation of p38 signaling. HuMSCs treatment significantly reduced the ratio of p-p38/p38 after HI modeling for 3 and 29 days (Fig. 7A and C). The ratio of p-p38/p38 at all experimental time points in the CYGB-HuMSCs group was significantly lower compared with the HuMSCs group (Fig. 7). Moreover, the ratio of p-p38/p38 in the CYGB-HuMSCs group was significantly lower compared with the sham group on day 3 after transplantation of CYGB-HuMSCs (Fig. 7A).

Figure 7. Expression levels of p-p38, p38, ERK, p-ERK, JNK and p-JNK after HI modeling for 3, 14 and 29 days (n=5 per group). GAPDH was used as the internal control. (A-C) The ratio of p-p38/p38 was significantly increased in the HIBD group after HI modeling for 3, 14 and 29 days. (A and C) The ratio of p-p38/p38 after HI modeling for 3 and 29 days in the HuMSCs group was decreased. (A-C) The ratio of p-p38/p38 at all experimental time points in the CYGB-HuMSCs group was significantly lower compared with the HuMSCs group. (B and C) The ratio of p-JNK/JNK in the HIBD group was significantly increased compared with the Sham group after HI modeling for 14 and 29 days. (B) On day 14 after HI modeling, compared with the HIBD group, the ratio of p-JNK/JNK was decreased in the HuMSCs and CYGB-HuMSCs groups, but those in the CYGB-HuMSCs group remained higher than the HuMSCs group. (C) On day 29 after HI modeling, there were no differences in the ratio of p-JNK/JNK between the HIBD group and the HuMCSs or CYgB-HuMSCs groups, but the ratio in the CYGB-HuMSCs group was significantly lower compared with the HuMSCs group. (A, B) The ratio of p-ERK/ERK in the HIBD group was significantly decreased compared with the Sham group after HI modeling for 3 days, while increased at day 14 after HI modeling. (A and C) On days 3 and 29 after HI modeling, compared with the HIBD group, the ratio of p-ERK/ERK was increased in the HuMSCs and CYGB-HuMSCs groups. (B) On day 14 after HI modeling, compared with the HIBD group, the ratio of p-ERK/ERK was decreased in the HuMSCs group, while increased in the CYGB-HuMSCs group. aP<0.01 vs. Sham group at the same time; bP<0.01 vs. HIBD group at the same time; cP<0.01 vs. HuMSCs group at the same time; a'P<0.05 vs. Sham group at the same time; c'P<0.05 vs. HuMSCs group at the same time; ns, no significant difference vs. other groups. p-, phosphorylated; HI, hypoxia-ischemia injury; CYGB, Cytoglobin; HuMSC, human umbilical cord-derived mesenchymal stem cells; HIBD, hypoxic-ischemia brain damage.

The ratio of p-JNK/JNK in the HIBD group was significantly increased compared with the Sham group after HI modeling for 14 and 29 days (Fig. 7B and C). On day 14 after HI modeling, compared with the HIBD group, the ratio of p-JNK/JNK was decreased in the HuMSCs group and CYGB-HuMSCs group; however, the ratio in the CYGB-HuMSCs group remained higher than the HuMSCs group (Fig. 7B). On day 29 after HI modeling, there were no significant differences in the ratio of p-JNK/JNK between the HIBD group and HuMCSs or CYGB-HuMSCs groups, but the ratio in the CYGB-HuMSCs group was lower compared with the HuMSCs group (Fig. 7C).

The ratio of p-ERK/ERK in the HIBD group was significantly decreased compared with the Sham group after HI modeling for 3 days, while increased at day 14 after HI modeling (Fig. 7A and B). On days 3 and 29 after HI modeling, compared with the HIBD group, the ratio of p-ERK/ERK was increased in the HuMSCs group and CYGB-HuMSCs group (Fig. 7A and C). On day 14 after HI modeling, compared with the HIBD group, the ratio of p-ERK/ERK was decreased in the HuMSCs group, while increased in the CYGB-HuMSCs group (Fig. 7B).

Discussion

HIBD is a significant threat to neonatal health, which results in disability and mortality in infants and children (5,6). Despite an increase in the number of studies that have been performed to investigate the pathogenesis of HIBD, hypothermia and systemic supportive treatment are the primary therapeutic options for HIBD (35). To the best of our knowledge, the present study was the first to demonstrate that intranasal transplantation of CYGB-HuMSCs exhibited neuroprotective and antiapoptotic effects in HIBD. In a rat model of HIBD, intranasal administration of both HuMSCs and CYGB-HuMSCs resulted in reduced severity of histological and functional deficits. In addition, the reduction in the CYGB-HuMSCs group was more prominent compared with the HuMSCs group.

CYGB has been extensively studied in the past 15 years (13), following its discovery. Different from hemoglobin, CYGB is found outside of red blood cells and is classified as a non-erythroid globin, similar to other globins, such as neuroglobin and myoglobin (36). The potential functions of these non-erythroid globins are associated with tissue and cell protection under hypoxic conditions, ischemic conditions and during oxidative stress (37). Due to the structure of the hemeglobin, CYGB can reversibly bind to oxygen and other small molecules (38). The ability of CYGB to store and sense oxygen, as well as its involvement in nitrite and NO metabolism, are being increasingly studied and understood (39,40). It has been reported that, similar to other hemoglobins, CYGB can scavenge ROS and NO, as well as produce NO from nitrite, thus reducing oxidative stress and protecting organs and cells (41–44). Our previous study revealed that pretreatment of CYGB-overexpression reduced neonatal rat HI injury, as well as improved long term memory and athletic ability following neonatal HI (18). However, whether overexpression of CYGB serves a protective role in neurons following HIBD has not been previously studied, and there is no established means of introducing the CYGB gene safely into living animals.

In the present study, the effect of transplantation of CYGB-HuMSCs in an HIBD rats model was determined, and was developed based the application of stem cells for treatment of various brain diseases and our previous study (18). The current results suggested that CYGB-HuMSCs reduced neuron cell apoptosis caused by HIBD. The improvement in long term neurological function 29 days after HIBD further demonstrated the neuroprotective effect of CYGB-HuMSCs transplantation. HuMSCs also exhibited a neuroprotective and antiapoptotic effect in HIBD, but the CYGB-HuMSCs exhibited more prominent neuroprotective effects. In addition, RT-qPCR and western blotting results identified that CYGB mRNA and protein expression levels were increased from day 3 to day 29 after transplantation of CYGB-HuMSCs. Thus, it was suggested that CYGB functions as an endogenous neuroprotective protein, and that CYGB-HuMSCs may act as a gene transfer vector for CYGB, allowing CYGB and HuMSCs to work together to improve the neuroprotective effects. Therefore, these results highlight a novel approach for treatment of neonatal HIBD.

Previous studies have shown that HuMSCs do not undergo cell replacement via differentiation into neuron-like cells, but rather exert neuroprotective effects (45,46). In the present study, HuMSCs or CYGB-HuMSCs administered via nasal transplantation did not differentiate into cells expressing neural or glial cell marker proteins; a result that is similar to previous studies (45,46). Therefore, it was speculated that the neuroprotective effects of HuMSCs may be achieved via the action of exosomes to promote endogenous repair of nerve cells (47–49). Proteins secreted by MSCs can promote the repair of damaged tissues via a variety of mechanisms, including prevention of apoptosis, regulation of inflammatory responses and promotion of endogenous repair mechanisms, such as angiogenesis and neurogenesis, rather than cell replacement (48). In a HIBD animal model, MSCs provide a suitable environment, including increasing the local levels of brain-derived neurotrophic factor, basic fibroblast growth factor, neurotrophic factor and nerve growth factor, whilst downregulating the expression of proinflammatory cytokines, such as IL-1 and IL-6, which may also contribute to the repair of nerve cells (47,50). Previous studies have reported that CYGB serves a neuroprotective role by reducing cerebral infarctions and apoptosis caused by oxidative stress in vivo, and also revealed that hypoxia inducible factor-1α/CYGB/vascular endothelial growth factor signaling may serve an important role in the CYGB-mediated antioxidant mechanism (18). Therefore, it was hypothesized that the neuroprotective effects of CYGB-HuMSCs is not via cell replacement, but via the various cytokines secreted by HuMSCs and the antioxidant effects of overexpression of CYGB transfected into animals by HuMSCs.

The p38 mitogen-activated protein kinase (MAPK) signaling pathway allows cells to interpret a wide range of external signals and respond appropriately by generating a plethora of different biological effects (51). It has been revealed that p38 MAPK is a mediator of hypoxia-induced cerebrovascular inflammation (52). The p38 MAPK signaling pathway enhances the production of a range of proinflammatory cytokines, including IL-β, TNF-α and IL-8 (53,54). Moreover, p38 MAPK has been suggested to be involved in the context of ischemia-induced stress in the brain (55). For example, p38 MAPK inhibits the hypoxia response pathway via EGL-9 in neurons (56). The p38 MAPK can also function as a mediator of ROS-mediated signaling, and either activate or suppress cell cycle progression, depending on the activation stimulus (57). Proteomic and biochemical analyses further demonstrated that p38 MAPK signaling mediates cell apoptosis (58), while in vitro studies have shown that overexpression of CYGB suppressed p38 expression and protected glomerular mesangial cells from oxidative stress (59).

In the present study, the ratio of p-p38/p38 was significantly increased in the HIBD group. However, HuMSCs supplementation protected against neuron cell apoptosis as demonstrated by the decreased ratio of p-p38/p38. In addition, the CYGB-HuMSCs group had an improved antiapoptotic effect accompanied with a greater degree of decreased p38 expression compared with the HuMSCs group. The present results also suggested that the CYGB-HuMSCs group had significantly downregulated phosphorylation levels of p38, which was accompanied with a significantly increased expression level of CYGB, compared with the sham group on day 3. Thus, it was indicated that exogenous CYGB may modulate p38 expression. The current results highlight the involvement of p38 MAPK signaling in HIBD, as well as suggesting that CYGB-HuMSCs treatment may protect animals from HIBD by avoiding activation of the p38 MAPK signaling pathway.

In conclusion, to the best of our knowledge, the present study was the first to demonstrate that intranasal transplantation of CYGB-HuMSCs can serve as a gene transporter, and that it exerts neuroprotective and antiapoptotic effects in HIBD. Additionally, it was identified that CYGB-HuMSCs may exert its protective properties via the p38 MAPK pathway. The results of the current study demonstrate a novel therapeutic approach for treatment of HIBD. However, further investigation is required to identify the mechanism via which CYGB-HuMSCs contributes to the molecular pathogenesis of HIBD.

Acknowledgements

Not applicable.

Funding

The present study was supported by the Science and Technology Project from the Science Technology and Innovation Committee of Shenzhen Municipality (grant no. JCYJ20160429141742207), the National Natural Science Foundation of China (grant nos. 81671525 and 81070478) and the Sanming Project of Medicine in Shenzhen (grant no. SZSM201512033).

Availability of data and materials

The datasets used and/or analyzed during the present study are available from the corresponding author on reasonable request.

Authors' contributions

SFT, YBC and HHY designed the study and performed the experiments. HHY, LM and LCX acquired the data. SFT and HHY analyzed the data. HHY, YBC and LM prepared the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

This study was approved by the Ethics Committees of Maternal and Child Health Care Hospital of Shenzhen University (Guangdong, China), and of the Institutional Animal Care and Use Committee of Shenzhen University (Guangdong, China). Written informed consent was obtained from the maternal donors and/or guardians.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References