The repair and autophagy mechanisms of hypoxia-regulated bFGF-modified primary embryonic neural stem cells in spinal cord injury.

Abstract There is no effective strategy for the treatment of spinal cord injury (SCI), a devastating condition characterized by severe hypoxia and ischemic insults. In this study, we investigated the histology and pathophysiology of the SCI milieu in a rat model and found that areas of hypoxia were unevenly interspersed in compressed SCI. With this new knowledge, we generated embryonic neural stem cells (NSCs) expressing basic fibroblast growth factor (bFGF) under the regulation of five hypoxia‐responsive elements (5HRE) using a lentiviral vector (LV‐5HRE‐bFGF‐NSCs) to specifically target these hypoxic loci. SCI models treated with bFGF expressed by the LV‐5HRE‐bFGF‐NSCs viral vector demonstrated improved recovery, increased neuronal survival, and inhibited autophagy in spinal cord lesions in the rat model due to the reversal of hypoxic conditions at day 42 after injury. Furthermore, improved functional restoration of SCI with neuron regeneration was achieved in vivo, accompanied by glial scar inhibition and the evidence of axon regeneration across the scar boundary. This is the first study to illustrate the presence of hypoxic clusters throughout the injury site of compressed SCI and the first to show that the transplantation of LV‐5HRE‐bFGF‐NSCs to target this hypoxic microenvironment enhanced the recovery of neurological function after SCI in rats; LV‐5HRE‐bFGF‐NSCs may therefore be a good candidate to evaluate cellular SCI therapy in humans.


| INTRODUCTION
Spinal cord injury (SCI) is a devastating event that usually results in significant functional impairment in the patient. SCI is a complex twostep process in which a cascade of secondary neurodegenerative events is set in motion by a primary injury. 1 Vascular changes, hypoxia, the loss of ATP-dependent processes, ionic disturbances, neurotransmitter accumulation, and apoptosis create a toxic milieu, perpetuating the SCI. In an attempt to contain the extent of the damage, the injured central nervous system (CNS) forms a glial scar-a potent physical barrier that prevents axonal regrowth through that region. 2,3 In an animal model, sustained compression of the spinal cord was shown to lead to a larger histologic lesion and poorer functional outcomes and inhibit the recovery of somatosensory-evoked potentials. 4 Effective strategies against this chronic phase of SCI have not been established; clinical trials of nonsurgical treatment options in human SCI have failed to demonstrate marked neurological benefit, in contrast to their success in the laboratory. A better understanding of the events of secondary injury would provide a target to optimize pharmacological and cellular therapies, the timing of surgery, and early rehabilitation.
The pathology behind the mechanism of secondary injury in SCI includes hypoxia and ischemia arising from impaired perfusion at the cellular level and the resulting cellular energy deficiency. 5 It has been reported that ischemia begins immediately after traumatic SCI and that if not treated, the spinal cord deteriorates in the first 3 hours and continues for at least 24 hours. 6 However, many studies have also indicated that the spinal cord has a remarkable healing capacity, and early revascularization and oxygenation of the injury are the most important factors to minimize the long-term, irreversible sequelae of SCI.
Reconstructive and regenerative experimental cellular strategies involving embryonic or adult stem cells or tissue, genetically modified fibroblasts, Schwann cells, olfactory ensheathing cells, and activated macrophages have been reported to exhibit varying degrees of recovery in different models of SCI. 5,7 A number of growth factors have also been shown to alter different cell types and functions, reducing the deleterious effects of an injury while improving neuronal survival and regeneration. 8 Angiogenic factors, such as basic fibroblast growth factor (bFGF), which is present in both neuronal and glial cells, have been proposed to address ischemia following injury and were previously reported to have multiple neuropromoting effects on the developing and adult nervous system of mice and other mammals. [9][10][11] The ischemic environment after SCI leads to limited neuron survival and complicates the transplantation of stem cells designed to promote a permissive environment at the implant site. 12,13 The implantation of exogenous bFGF or its overexpression in mesenchymal stem cell therapy can spare spinal cord tissue, reduce retrograde degeneration, improve vascularity, and reduce the number of apoptotic cells. 14,15 However, controlling the release of these factors is a significant challenge because of the potential for increased microvascular permeability associated with an increase in lesion volume and neoplastic development secondary to uncontrolled cell differentiation. 16 To the best of our knowledge, the delivery of embryonic neural stem cells (NSCs) with the controlled expression of bFGF under the regulation of five hypoxia-responsive elements (5HRE) using a lentiviral vector (LV-5HRE-bFGF-NSCs) to specifically target hypoxia, and the ischemic microenvironment in SCI has not yet been investigated.
In this study, a rodent model of compressive SCI was established.
The expression patterns of hypoxia-inducible factor-1α (HIF-1α), a transacting factor widely expressed in ischemia, were studied at the mRNA and protein levels, and we discovered a pattern of hypoxia with increased cell autophagy throughout the injured spinal cord. 17 Furthermore, our study shows that the application of LV-5HRE-bFGF-NSCs to specifically target these hypoxic loci reversed the hypoxic microenvironment at day 60 after SCI, concomitant with decreased cellular autophagy, reduced CNS glial scar formation, and improved locomotor function in in vivo studies. The results of this study increase the current understanding of the pathophysiology of SCI and may be employed to combat the ischemic microenvironment, which can induce cell death and limit cell transplantation approaches, to promote spinal cord regeneration.

Significance statement
The present study shows that application of hypoxiaregulated basic fibroblast growth factor modified primary embryonic neural stem cells to specifically target the hypoxic loci resulted in a reversal of the hypoxic microenvironment after spinal cord injury (SCI), concomitant with decreased cellular autophagy, reduced CNS glial scar formation, and improved locomotor function in in vivo studies.
The results of the present study increase the current understanding of the pathophysiology of SCI and may be used to combat the ischemic microenvironment that can induce cell death and limit cell transplantation approaches to promote spinal cord regeneration. chemiluminescence kit and CM-DiI were purchased from Bio-Rad (Hercules, California). Thapsigargin (TG) and 3-methyladenine (3-MA) were purchased from Sigma-Aldrich. The autophagy activator rapamycin (RAPA) was purchased from Cell Signaling Technology. All other reagents were purchased from Beyotime Institute of Biotechnology (Shanghai, China) unless otherwise specified.

| Isolation and culture of rat embryonic derived NSCs
Wistar rat embryos were obtained at 14-16 days after pregnancy by routine surgical procedure, followed by separating and immersing the cerebral cortex of each embryo in D-Hanks solution. The cerebrovascular and meninges were carefully removed under the microscope and centrifuged at 1000 rpm for 3 minutes, the cerebral cortex was then shredded and treated with trypsin (0.125%) and EDTA (0.102%). 18 Digestion with trypsin/EDTA was terminated by culture medium DMEM/F12 containing 10% FBS, 1% N2 supplements, 2% B27, 20 ng/mL bFGF, 20 ng/mL EGF, 200 IU/L penicillin, and 100 IU/L streptomycin, followed by collecting and centrifuging the cell suspension. 18 After resuspension with culture medium, cells were transferred into T25 cell culture flasks (0.5 × 10 6 cells/flask) previously coated with laminin and poly-ornithine and cultured at 37 C in a moist atmosphere containing 5% CO 2 .

| Cell culture and preparation
NSCs were maintained in DMEM supplemented with 10% FBS and 5% horse serum with 2 mM glutamine and penicillin/streptomycin. The cells were cultured in a humidified atmosphere containing 5% CO 2 and 95% air at 37 C and passaged at 90% confluence. To further evaluate the effect of TG on the survival and proliferation of NSCs, NSCs were treated with TG for 12 hours as previously described.
According to our previous study, cells were treated with RAPA (100 nM) or the autophagy inhibitor 3-MA (5 mM) to further determine the role of autophagy on primary NSCs at 12 hours. All experiments were performed in triplicate.

| Lentiviral transduction
NSCs were seeded in 24 wells plate at a density of 2 × 10 5 cells per well. NSCs were exposed to the viral particles in 0.5 mL DMEM at 37 C medium for 4 hours. Cells were transduced with LV-GFP, LV-bFGF, LV-5HRE-GFP, and LV-5HRE-bFGF using polybrene (a final concentration of 8 μg/mL). The medium was then removed, and the cells were washed once with DMEM and then re-cultured with normal medium with Blasticidin (2 μg/mL) for 14 days. The untransduced cells were eliminated after 14 days by culturing with Blasticidin. Following the stable selection, we found that GFP signal initially observed after transduction was barely detected by confocal analysis, suggestive that IRESs in these constructs were not active in NSCs under the experimental conditions described. Nevertheless, the expression of bFGF was subsequently confirmed by Western blot and confocal analyses as well as enzyme-linked immunosorbent assay (ELISA).

| Animal model of SCI
Eight-week-old female Sprague-Dawley rats weighing 220-250 g were purchased from the Animal Center of Chinese Academy of Sciences, Shanghai, China. Animals were housed for at least 7 days before the experiment in a room with a 12-hour light/dark cycle at 23 C-25 C and received free access to water and food. All protocol of the animal use and care was conducted according to the Guide for the Care and Use of Laboratory Animals from the National Institutes of Health and was approved by the Animal Care and Use Committee of Wenzhou Medical University. All experiments conformed to named local and international guidelines on the ethical use of animals. All the animals were anesthetized by an intraperitoneal injection of 10% chloralic hydras (3.5 mL/kg). Rats were then positioned on a cork plat- form. An incision in the epidermis along the midline of the back was performed to expose the vertebral column, and a laminectomy was performed at T9 segment of the spine. Moderate crushed injuries were compressed by a vascular clip for 2 minutes (30 g forces, Oscar, China). 15 Control group animals received the same surgical procedures without impaction. Postoperative nursing contained the artificial emptying of the bladder, twice a day, until the rats restored their bladder function by using cefazolin sodium (50 mg/kg, i.p.).

| Transplantation
The animals received transplantation 1 week after SCI. This time point is widely accepted as a suitable therapeutic window as the inflammatory reaction (creating a hostile environment for cell transplant survival) decreases during the first 7 days, and the glial scar that prevents graft host tissue communication is not yet developed. 19 The animals were fixed in a stereotaxic instrument with a rat-specific vertebra-holder (Cunningham spinal adaptor, Stoelting Co., Wood Dale, Illinois), receiving exposure at T9 of spine. A total of 1 × 10 6 NSCs cells /5 μL (either unlabeled or labeled with CM-DiI) were injected through 10 μL microinjector (26G, an inner diameter of 0.24 mm, an outer diameter of 0.6 mm, 30 bevel, 1-cm long needle) into the epicenter at a depth of 1 mm below the dorsal surface at a rate of 1 μL/min using a Nano-Injector (Stoelting Co.). Cells were obtained as described above, and the culture medium suspension was prepared before transplantation. The number of cells was determined based on our pilot study, in which 2 × 10 5 NSCs cells /1 μL were also injected into the proximal, central and distal parts of the injured spinal cord. The microinjector was kept in place after injection for another 5 minutes to prevent cell suspension leakage.
The control group received 5 μL of phosphate-buffered saline (PBS).

| Behavioral recovery evaluation
In order to define recovery features after SCI, behavioral analyses were conducted by trained investigators who were blind to the experimental conditions. Basso-Beattie-Bresnahan (BBB) locomotion scale, inclined plane test, and footprint were performed as described else- where to evaluate open-field locomotion. 20  The inclined plane test was performed by a testing apparatus. 21 The maximum angle at which a rat could retain its position for 5 seconds without falling was recorded for indicated position and averaged to obtain a single score for each animal. Footprint analysis was conducted through dipping the animal's hind limbs into blue dye as described elsewhere, 21,22 followed by recording and analyzing walking path through a narrow box (1 m long and 7 cm wide).
The average of six values was used for statistical evaluation. The data are expressed as mean ± SEM. All data were compared between the sham operated group and the transplanted group by a two sample ttest for independent samples, if the two samples had equal variances. If they had unequal variances, the Mann-Whitney test was used for evaluation. A P-value <.05 was considered statistically significant. All behavioral tests were performed by two independent blind observers.

| Hematoxylin-eosin staining and Nissl staining
The rats were anesthetized with 10% chloral hydrate (3.5 mL/kg, i.p.) at 1 month and 2 months after injury, and perfused with 0.9% NaCl, followed by perfusing with 4% paraformaldehyde in 0.01 M PBS (pH = 7.4). The spinal cords from the C1-L5 segments were excised and fixed in 4% paraformaldehyde overnight at 4 C, and then embedded in paraffin. Transverse paraffin sections (5 μm thick) were mounted on poly-L-lysine-coated slides for hematoxylin-eosin (H&E) staining or Nissl staining. Images were captured by a light microscope.

| Immunofluorescence staining
Animals were perfused through the heart with cold saline followed by washed four times with PBS containing 0.1% Triton X-100 at room temperature for 10 minutes each time, followed by three times with PBS for 5 minutes each time and briefly with water. All images were captured by Nikon ECLIPSE Ti microscope (Nikon, Tokyo, Japan).

| Apoptosis assay
The

| Statistical analysis
All data were expressed as the mean ± SEM. Statistical significance was determined with Students' t test when there were two experimental groups. For more than two groups, statistical evaluation of the data was performed using one-way analysis of variance test, followed by Dunnett's post hoc test. Values of P < .05 were considered significant. We chose bFGF because bFGF is a known neuroprotectant against a number of brain and SCI conditions.
We generated NSCs expressing bFGF using a lentivirus system.
Under noninducible conditions ( Figure S1A), LV-bFGF-NSCs exhibited higher expression levels of bFGF than LV-GFP-NSCs and NSCs ( Figure S1B-D), and bFGF expression levels increased steadily over time ( Figure S1E). Under an inducible expression system ( Figure 1A 1%). Especially at low concentrations or free serum conditions, the proliferation rate is more significant in LV-5HRE-bFGF-NSCs. * represents P < .05 and ** represents P < .01 vs the 1d group, # represents P < .05 and ## represents P < .01 vs NSCs and LV-5HRE-GFP-NSCs group. Data are the mean values ± SEM. F-G, MTT assay results of different experimental groups treated with TG for 12 hours in vitro. * represents P < .05 or ** represents P < .01, # represents P < .05. Data are the mean values ± SEM. All experiments were repeated three times. 5HRE, five hypoxia-responsive elements; bFGF, basic fibroblast growth factor; NSC, neural stem cell 3.4 | LV-5HRE-bFGF-NSCs showed expanded axon regeneration over the scar boundary, accompanied by astrocyte scar inhibition and increased NSC migration, proliferation, and differentiation We then examined the expression of growth associated protein 43 (GAP43), a crucial component of the axon that plays a key role in axonal regeneration and plasticity. 24,25 GAP43 was expressed at the highest level in the LV-5HRE-bFGF-NSCs group on days 14 and 60 ( Figure 5A-L) after transplantation. By confocal analysis, we observed that GAP43 expression signals (green) were detected in almost >90% of transplanted NSCs (red plus green) and some residual NSCs (green only). Notably, GAP43-positive transplanted NSCs (yellow) migrated beyond the scar (white) from the injury site to normal regions ( Figure 5A and G), and this migration was most obvious in the LV-5HRE-bFGF-NSCs group.
Finally, cross sections of the SCI were examined histologically, and CM-DiI labeling allowed the migration of NSCs from the initial transplantation site to be easily observed ( Figure 6A-C and Figure S5A-C). More NSCs in the LV-5HRE-bFGF-NSCs group survived, and NSCs in the 5HRE-bFGF-NSCs group migrated better than NSCs in the other groups.
These results indicate that the administration of LV-5HRE-bFGF-NSCs  improved, bFGF expression then returns to its normal levels.

| DISCUSSION
With our increased understanding of the hypoxic SCI milieu, we undertook further investigations to increase the target specificity and improve the safety and efficacy of the expression system used to combat SCI. Animals received transplanted cells 7 days after SCIafter the initial inflammatory infiltration had decreased and bFGF upregulation had peaked and prior to formation of the glial scar that prevents graft-host tissue communication. 19 Remarkably, hypoxic conditions were most prominent on day 14 after SCI, as shown by the highest level of HIF-1α expression. This was accompanied by increased glial scar formation. The combination of NSC transplantation with bFGF expression in the right place at the right time will likely reduce glial scar formation.
The following several lines of evidence support the observation that LV-5HRE-bFGF-NSCs assisted in the regeneration of neurons after SCI in this study: 1. LV-5HRE-bFGF-NSCs reversed the hypoxia-induced milieu of SCI at day 42 after transplantation and returned the microenvironment to conditions of normoxia at day 60.
2. LV-5HRE-bFGF-NSCs were superior in protecting against cellular apoptosis and autophagy in vitro and in vivo, induced cell proliferation under serum deprivation, and optimized the differentiation of NSCs.
3. Histological analysis showed that CM-Dil-labeled NSCs in the LV-5HRE-bFGF-NSCs groups migrated across the scar boundary.

A marked improved in the anatomical appearance of injured spinal
cords after treatment with LV-5HRE-bFGF-NSCs was associated with superior locomotor functional recovery. Importantly, we failed to observe the same level of neuron regeneration and locomotor function restoration unless each of the above mechanisms that individually contribute to SCI degeneration was targeted.

| CONCLUSION
We report the novel finding of unevenly interspersed hypoxic conditions with increased cell autophagy throughout the injury site of SCI in rats. Established injuries require a better appreciation of altered neuropathophysiology, which will be a determining factor in the restoration of function consequent to successful neuron regeneration by cellular strategies. In this study, we constructed LV-5HRE-bFGF-NSCs, which were successful in simultaneously targeting multiple neural mechanisms responsible for regeneration failure. Improved functional outcomes resulting from neuron regeneration will most likely require advances in each of these mechanisms.

DATA AVAILABILITY STATEMENT
The data used to support the findings of this study are available from the corresponding author upon request.