Transplantation of fibrin-thrombin encapsulated human induced neural stem cells promotes functional recovery of spinal cord injury rats through modulation of the microenvironment

Abstract

Recent studies have mostly focused on engraftment of cells at the lesioned spinal cord, with the expectation that differentiated neurons facilitate recovery. Only a few studies have attempted to use transplanted cells and/or biomaterials as major modulators of the spinal cord injury microenvironment. Here, we aimed to investigate the role of microenvironment modulation by cell graft on functional recovery after spinal cord injury. Induced neural stem cells reprogrammed from human peripheral blood mononuclear cells, and/or thrombin plus fibrinogen, were transplanted into the lesion site of an immunosuppressed rat spinal cord injury model. Basso, Beattie and Bresnahan score, electrophysiological function, and immunofluorescence/histological analyses showed that transplantation facilitates motor and electrophysiological function, reduces lesion volume, and promotes axonal neurofilament expression at the lesion core. Examination of the graft and niche components revealed that although the graft only survived for a relatively short period (up to 15 days), it still had a crucial impact on the microenvironment. Altogether, induced neural stem cells and human fibrin reduced the number of infiltrated immune cells, biased microglia towards a regenerative M2 phenotype, and changed the cytokine expression profile at the lesion site. Graft-induced changes of the microenvironment during the acute and subacute stages might have disrupted the inflammatory cascade chain reactions, which may have exerted a long-term impact on the functional recovery of spinal cord injury rats.

Introduction

Spinal cord injury (SCI) leads to interruption of ascending and descending neural circuits and results in severe functional impairment. Currently, there is no effective clinical treatment to cure SCI. In the research field, effort is being focused on development of novel treatment strategies including transplantation of biomaterials, growth factors, and cell therapies (Butts et al., 2017; Führmann et al., 2017; Khazaei et al., 2017; Santhosh et al., 2017; Shroff et al., 2017; Li et al., 2018; Hu et al., 2019). Certain biomaterials have been tested in animal models, with or without cellular grafts, to act as bridging scaffolds and fill in the gap at the lesion site. Chitosan (glucosamine [1–4]-2-amino-b-d glucose) has been well investigated and can facilitate the accommodation, growth, and differentiation of cellular grafts (Gao et al., 2014). Poly lactic-co-glycolic acid combined with mesenchymal stem cells (MSCs) showed a better outcome when transplanted into a SCI model, compared with transplantation of MSCs alone (Yousefifard et al., 2019). Collagen sponge used as a narrow bridging tissue can connect spinal cord stumps and restore motor function following SCI (Yoshii et al., 2004; Han et al., 2015; Li et al., 2017). Thrombin and fibrinogen have been used due to their favorable biocompatibility, with fibrinogen-specific antibody used to detect fibrin (Sharp et al., 2012) and examine the kinetics of the biomaterial. Fibrinogen is produced by the liver and can form an insoluble reticular structure under catalyzation of thrombin, a key process involved in blood clotting. Researchers have further employed fibrinogen combined with thrombin in the treatment of SCI. In a rhesus monkey model, fibrin-thrombin loaded human neural progenitor cells (NPCs) engrafted into sites of cervical SCI matured into neurons, extended axons, and formed synapses with host cells (Rosenzweig et al., 2018).

Transplantation of cellular grafts alone have also demonstrated certain beneficial effects in SCI animal models. Neural stem cells (NSCs) engrafted into NOD-scid mice (immunodeficient mice on a non-obese diabetic background) differentiate into neurons and promote functional recovery following SCI (Ogawa et al., 2002; Cummings et al., 2005; Liu et al., 2017). Transplanted oligodendrocyte precursor cells were able to improve recovery of motor function following contusion-induced SCI in rats (Manley et al., 2017; Yang et al., 2018). Following SCI, the spinal microenvironment is disrupted, leading to a series of pathophysiological changes such as macrophage activation (Gensel and Zhang, 2015), downregulation of beneficial factors, and upregulation of harmful factors such as neurotrophic factors, cytokines, and chemokines (Fan et al., 2018). Imbalance of the SCI microenvironment can impair neural regeneration and functional recovery. MSCs have been tested in several disorders, including SCI, and may modulate the microenvironment at the lesion site (Park et al., 2012; Fu. et al., 2017). MSC grafts improved the SCI microenvironment by reducing expression of interleukin (IL)-6 and IL-17a and enhancing the expression of transforming growth factor-β (TGF-β), resulting in promotion of functional recovery of SCI rats.

Studies have shown that NSC engraftment enhanced differentiated neurons, astrocytes, and oligodendrocytes, as effectors to facilitate recovery (McMillan et al., 2022; Zhang et al., 2022). However, few studies have used NSCs and/or biomaterials as major modulators of the SCI microenvironment. Fibrin-thrombin scaffolds have been used to treat disorders such as wound healing (Borchers and Pieler, 2010) and glioblastoma by accommodating chimeric antigen receptor (CAR)-T cells (Ogunnaike et al., 2021) because they do not cause inflammation or tissue necrosis. In the current study, we transplanted fibrin-thrombin encapsulated human induced NSCs (iNSCs) reprogrammed from peripheral blood mononuclear cells (PBMCs), into a rat SCI model to investigate the impact of transplantation on the microenvironment of the injured spinal cord.

Methods

Animals and ethics statement

Healthy female Sprague-Dawley rats (210–230 g, 8–10 weeks old, n = 81) were chosen for modeling complete SCI and transplantation because they are easier to care for post-operation and are less aggressive than injured male rats (Robinson and Lu, 2017). The rats were specific pathogen-free (SPF) grade, purchased from Vital River (Beijing, China, license No. SCXK (Jing) 2021-0006) and housed in temperature- and humidity-controlled animal quarters (temperature of 22–25°C, relative humidity of 50–65%) with a 12-hour light/dark cycle. All rats were allowed free access to food and water at all times and were kept two per cage. The rats used for all experiments were naïve and no drug tests were done. All animal experiments were performed in accordance with the Chinese Ministry of Public Health Guide and US National Institutes of Health Guide for the care and use of laboratory animals. All procedures performed in studies involving animals were approved by the Animal Ethics Committee of Xuanwu Hospital Capital Medical University (XW-20210423-2, approval date April 23, 2021) and is reported in accordance with the ARRIVE 2.0 guidelines (Animal Research: Reporting of In Vivo Experiments) (Percie du Sert et al., 2020).

Cell culture and preparation

Human induced neural stem cells were generated as previously reported (Yuan et al., 2018). Briefly, human PBMCs were isolated from the blood of two healthy men (age 25 and 28 years; 30 mL blood each) under informed consent. Cultured PBMCs were then infected by Sendai virus (Life Technologies, Carlsbad, CA, USA) encoding OCT3/4, SOX2, KLF4, and c-MYC (OSKM) (multiplicity of infection [MOI] = 10). After 2 days of infection, cells were changed into iNSC proliferation medium consisting of DMEM/F12 (Gibco, Carlsbad, CA, USA), Neurobasal-A (Gibco), N2 (50×, Gibco), B27 (100×, Gibco), GlutaMAX (100×, Gibco), NEAA (100×, Gibco), CHIR99021 (final concentration 3 μM), SB431542 (final concentration 2 μM), and leukemia inhibitory factor (LIF) (final concentration 10 ng/mL). Equal volumes of DMEM/F12 and Neurobasal-A were used. The medium was prepared and stored at 4°C for 1 week. Prior to changing, the medium was placed at room temperature instead of 37°C. The medium was half-changed every 2 days and passaged every 4–6 days. Prior to cell transplantation, iNSCs were dissociated into single cells with Accutase (Gibco) at 37°C for 5–20 minutes, followed by centrifugation at 250 × g for 5 minutes, and then mixed with the biomaterial (described in “Animal surgery and transplantation”).

Biomaterial preparation and cell loading

Human fibrinogen (100 mg/mL, final concentration 25 mg/mL; Cat# F3879, Sigma, St. Louis, MO, USA) was dissolved with 20 mM CaCl₂-water solution. Human thrombin (100 U/mL, final concentration 25 U/mL; Cat# T7009, Sigma) was dissolved with 0.1% bovine serum albumin (BSA). Human iNSCs were collected into 1.5 mL Eppendorf tubes, resuspended with 50 μL 0.1% BSA (50 μL BSA-iNSCs), and then thoroughly mixed with 50 μL thrombin (100 U/mL) (100 μL BSA-iNSCs-thrombin). Meanwhile, 50 μL human fibrinogen (100 mg/mL) was dissolved in 50 μL 20 mM CaCl₂ (100 μL fibrinogen-CaCl₂), and quickly mixed (within 3 seconds) with 100 μL BSA-iNSCs-thrombin. Fibrin-thrombin iNSCs spontaneously cross-link into a gel-like soft mixture and were warmed in an incubator for 3–5 min prior to transplantation (described in “Animal surgery and transplantation”).

Animal surgery and transplantation

All animals within an experimental group subjected to SCI were randomly assigned to receive either neural grafts loaded in biomaterial, biomaterial only, or lesion only. A total of 81 rats were used. Of these 81 rats, 6 were randomly assigned to the normal group without being subjected to SCI. The remaining 75 rats were randomly assigned to SCI groups: SCI only group, SCI + material alone group, and SCI + iNSCs transplantation group. Among the 75 SCI rats, 62 rats survived injury. The SCI only group consisted of 22 rats and included 5 days post-injury (dpi), 15 dpi, 30 dpi, 60 dpi (n = 3/time point), and 7 months pi (7 months pi, n = 10). The SCI + material alone group consisted of 19 rats, including 5 dpi, 15 dpi, 30 dpi, 60 dpi (n = 3/time point), and 7 months pi (n = 7). The SCI + iNSCs transplantation group consisted of 21 rats, including 5 dpi, 15 dpi, 30 dpi, 60 dpi (n = 3/time point), and 7 months pi (n = 9).

The SCI model used involved a complete transection at thoracic (T)8–T9. For complete SCI, the animals were deeply anesthetized by intraperitoneal injection of a combination (2 mL/kg) of ketamine (25 mg/mL, Gutian, Fujian, China), xylazine (1.3 g/mL, Sigma), and acepromazine (0.25 mg/mL, Sigma). A 2 cm midline incision was made to expose the T8–T11 vertebrae. The spinal cord was exposed following resection of the laminae. A 2 mm long gap was introduced by sectioning the spinal cord between T8 and T9, with removal of the spinal segment. Bleeding was controlled using a hemostatic sponge in the transection site. The musculature and skin were sutured, after which all rats were returned to their home cages for recovery.

Prior to transplantation, biomaterial alone or cell-loaded biomaterial were gently but thoroughly mixed and cross-linked to produce a soft clot. Biomaterial (with or without cells) were maintained at a total volume of 200 μL, which was sufficient to fill the lesion gap. For the iNSC transplantation group, 4 × 10⁶ cells loaded in biomaterial were grafted into the lesion center (T8–T9) of the spinal cord for each rat. Then, 4 × 10⁶ fibrin-thrombin encapsulated iNSCs were transplanted into the lesion core of the 2 mm long injured spinal cord of each rat. For each rat, prior to transplantation, 4 × 10⁶ dissociated iNSCs were resuspended with 50 μL 0.1% BSA (50 μL BSA-iNSCs), which was thoroughly mixed with 50 μL thrombin (100 U/mL) (100 μL thrombin-BSA-iNSCs). Meanwhile, 50 μL human fibrinogen (100 mg/mL) was dissolved in 50 μL 20 mM CaCl₂ (100 μL fibrinogen-CaCl₂), which was quickly mixed (within 3 seconds) with 100 μL thrombin-BSA-iNSCs. Fibrin-thrombin iNSCs spontaneously cross-link into a gel-like soft mixture. For the material alone transplantation group, the same procedure as described above was conducted except that 50 μL 0.1% BSA solution without cells was used. The iNSC-material or material alone grafts were placed into the lesion core immediately following SCI. After transplantation, all rats were treated with the antibiotic ampicillin (Keda, Jiangxi, China) for 1 week. All animals (SCI only, SCI + material alone, SCI + iNSCs transplantation) were treated with cyclosporine D to avoid rejection of the iNSC graft. Daily injections of cyclosporine D (10 mg/kg) began one week before transplantation until sacrifice of the animals. The rats’ bladders were manually voided twice daily following SCI until euthanasia. The study flowchart and timeline are shown in Additional Figures 1 ^{(574.5KB, tif)} and 2 ^{(289KB, tif)} .

Tissue collection

SCI rats were euthanized at different time points post-injury and the spinal cord was sectioned and stained for analysis. The rats were anaesthetized with a combination (2 mL/kg) of ketamine (25 mg/mL), xylazine (1.3 g/mL), and acepromazine (0.25 mg/mL). Rats were transcardially perfused with ice-cold 0.9% saline. The spinal cord from T6–T10 was dissected, fixed in 4% paraformaldehyde (PFA) for 48 hours at 4°C, and then transferred to 20% and 30% sucrose for 24 hours each. The T6–T10 segments were then embedded in optimal cutting temperature compound (OCT), cut into 20 μm thick sections (Leica Microsystems, Wetzlar, Germany), and stored at –80°C.

Immunofluorescence staining and quantification

Every fifth section of each rat was collected for staining (n ≥ 3 rats/group). The sections were treated with 0.3% Triton X-100 in phosphate-buffered saline (PBS) for 10 minutes, and blocked with 10% BSA for 1 hour at room temperature. The sections were then incubated with primary antibodies (1:500 dilution) at 4°C overnight. The antibodies are listed in Additional Table 1. The slides were washed three times with PBS and subsequently incubated with conjugated secondary antibodies (Invitrogen, Waltham, MA, USA) for 2 hours at room temperature. 4′,6-Diamidino-2-phenylindole (DAPI) (1 mg/mL; YEASEN, Shanghai, China) was used to counterstain nuclei. Images were captured using a confocal microscope (Leica SCN400 Slide Scanner, Leica Microsystems). For quantification of positive cells with typical marker expression patterns, at least nine fields of each section at 200 × magnification were sampled and analyzed using ImageJ software 1.51 (U. S. National Institutes of Health, Bethesda, Maryland, USA; Schneider et al., 2012). The average positive cell percentage (number of positive cells/number of total cells) was calculated per rat sample as a replication sample for statistics. The average percentage of positive cells was calculated.

Additional Table 1.

Antibodies used in immunofluorescence staining

Antibody	Species	Dilution	Cat#	Company
BDNF	Rabbit	1:500	bs-4989R	Bioss, Beijing, China
CD206(marker for M2 macrophage)	Mouse	1:500	60143-1-lg	Proteintech, Chicago, IL, USA
CD45 (marker for immune cells)	Mouse	1:500	60287-1-lg	Proteintech
CD68 (marker for immune cells)	Mouse	1:500	28058-1-AP	Proteintech
CD86 (marker for M1 macrophage)	Mouse	1:500	13395-1-AP	Proteintech
GFAP (marker for astrocytes)	Chicken	1:500	ab4674	Abcam, Cambridge, UK
Human fibrinogen	Mouse	1:500	bsm-1240M	Bioss
Human nuclei (Hu)	Mouse	1:400	MAB1281	Millipore, Billerica, MA, USA
Iba1 (marker for microglia)	Goat	1:500	ab955	Abcam
Laminin	Rabbit	1:500	L9393	Sigma, St. Louis, MO, USA
Map2(marker for neurons)	Rabbit	1:500	ab32454	Abcam
NF200 (marker for neurons)	Mouse	1:500	MAB5262	Millipore
Sox2 (marker for NSCs)	Goat	1:500	GTX101507	GeneTex, San Antonio,TX, USA
Syn (marker for synapses)	Mouse	1:500	Ab1543	Millipore
TGFβ	Rabbit	1:400	YT4632	Immunoway, Plano, Texas, USA
TGFβR	Rabbit	1:400	Ab31013	Abcam
TNFα	Rabbit	1:400	YT4689	Immunoway
TNFαR	Rabbit	1:400	21574-1-AP	Proteintech
Tuj1(marker for neurons)	Rabbit	1:500	MAB1637	Millipore

Behavioral assessment

Functional recovery was evaluated weekly throughout the study. The first test was performed 7 days (1 week) after the operation, and thereafter at 7 day intervals for 28 weeks. Functional analysis was performed in each group by two observers blinded to treatments. Recovery of hind limb locomotor function after SCI was assessed in an open field with a diameter of 2 m, using the Basso, Beattie and Bresnahan (BBB) locomotor rating scale (Basso et al., 1995). Each rat received a continuous 5 minute double-blind assessment of motion characteristics. The BBB score is divided into 0–21 levels based on motor coordination of the rats. A score of 21 indicates normal function, and a score of 0 indicates complete loss of function. The higher the score, the better the recovery of motor function of the rat’s hind limbs.

Motor evoked potential examination

Electrophysiological study was conducted for each group (n = 6 rats/group). Before examination, the animals were anesthetized with a combination (2 mL/kg) of ketamine (25 mg/mL), xylazine (1.3 g/mL), and acepromazine (0.25 mg/mL). Latency and amplitude of motor evoked potential (MEP) were then measured using Keypoint-II bi-channel evoked potentials/electromyograph (Dantec Dynamics, Skovlunde, Denmark). The stimulating electrodes (needle electrodes) were subcutaneously inserted into the cerebral motor cortex at the intersection of the sagittal suture and both external auditory meatuses, according to the rat brain atlas (Paxinos and Watson, 1998). Recording electrodes were inserted into the gastrocnemius muscle of the contralateral hind limb.

Lesion volume measurement

Lesion volume measurement was performed as previously reported (Sharp et al., 2012). To define the lesion more accurately, the volume was calculated from glial fibrillary acidic protein (GFAP)-stained sections. Specifically, the area of three manual traces per section were averaged, with the areas from 20 to 25 consecutive sections collected at a defined spacing per cord and summed. The known distance between each section was used to calculate the lesion volume of each cord in cubic millimeters.

Histological analysis

The collected tissue (T6–T10) was fixed in 4% PFA for 48 hours, embedded with OCT and stored at –80°C. OCT-embedded frozen tissue was cut into 20 μm thick frozen sections. Adjacent tissue sections were stained with hematoxylin and eosin (H&E) (Beyotime, Shanghai, China) for general observation. Luxol Fast Blue (LFB) staining was used to identify myelin in regenerated nervous tissue, according to the instructions of a LFB kit (Solarbio, Beijing, China). Images were captured using a Pannoramic Scanner (3DHISTECH, Budapest, Hungary).

Statistical analysis

No statistical methods were used to predetermine sample sizes; however, our sample sizes are similar to those reported in previous publications (Lu et al., 2012; Li et al., 2017; Robinson and Lu, 2017). All experiments and analyses were conducted with the investigators blinded to experimental conditions. Relative expression of positive cells and BBB scores were calculated using GraphPad Prism 5 (GraphPad Software, San Diego, CA, USA, www.graphpad.com). The data are presented as mean ± SD. Three or more groups were compared using two-way analysis of variance with Sidak’s post hoc test or one-way analysis of variance with Tukey’s post hoc test. P value < 0.05 was considered as significant.

Results

Short-term survival of graft

In our previous study, we successfully converted human PBMCs into iNSCs (Yuan et al., 2018). Importantly, in vivo transplantation of either iNSCs or iNSC-derived dopaminergic precursors into the brain of immunodeficient mice did not cause tumor formation (Yuan et al., 2018). Here, iNSCs were expanded in proliferation medium for several passages, dissociated into single cells, and then mixed with fibrinogen and thrombin to form a soft clot. This clot was then transplanted into the lesion center of the completely transected spinal cord (Figure 1A and B and Additional Figure 3 ^{(426.1KB, tif)} ). Once mixed, the material and cells solidify rapidly (usually within 3 seconds) at room temperature. Five days post-injury, Hu⁺ cells (stained for human nuclei, a marker of human cells) were detected at the lesion site (Figure 1C and E). Some grafted cells (36% of total grafted cells) were also observed beyond the lesion site, mostly in areas caudal to the lesion boundary, at a distance up to 2 mm (Figure 1C and D). In addition, Hu⁺ cells were co-labeled with the NSC marker, Sox2, but were negative for GFAP, a marker for astrocytes that usually stains negative at the lesion core and can be used to define the lesion boundary (Figure 1E). However, Hu⁺ cells failed to be detected by day 15 onward after injury (data not shown). These results suggest that the transplanted iNSCs could survive at least short-term in the injured spinal cord.

Figure 1.

Graft survival 5 days after transplantation into the spinal cord of SCI rats.

(A) Schematic representation of the experimental procedure. (B) Establishment of a complete SCI model. (1) Exposure of spinal cord on T8–T9 after resection of the laminae. (2) Removal of 2 mm long spinal tissue. (3) Suture of the musculature. (4) Suture of the skin and application of betadine. (C) Schematic diagram of iNSC migration within the injured spinal cord at 5 days after transplantation. Induced neural stem cells (iNSCs), green dots; biomaterial, red lines. (D) Transplanted iNSCs express human nuclear antigen (Hu) at and around the lesion site at 5 days post-injury (dpi). n = 3. 4′,6-Diamidino-2-phenylindole (DAPI in blue) indicates host and graft cell nuclei. Hu (in green) indicates nuclei of engrafted human iNSCs. (E) Transplanted iNSCs in the lesion center co-express Hu and Sox2 (a marker of NSCs) at 5 dpi. n = 3. Stars indicate biomaterial. Blue, DAPI; Green, Hu; Red, glial fibrillary acidic protein (GFAP); White, Sox2. Scale bar: 100 μm. SCI: Spinal cord injury.

Transplantation of iNSCs and biomaterial promotes functional recovery after SCI

A certain level of spontaneous recovery would normally occur following SCI, as shown by the BBB scoring results (Figure 2A). Compared with the SCI only group, engraftment of material alone significantly augmented the locomotor function of SCI rats (Figure 2A). Transplantation of iNSCs together with material further enhanced locomotor function (Figure 2A). To examine the neuronal circuitry that controls locomotor function, we performed electrophysiological analyses by placing a stimulating electrode at the motor cortex and a recording electrode at the gastrocnemius muscle of the hindlimb. We recorded electrophysiological signals at both the left and right hindlimb. In naïve rats without SCI, electrical stimulation at the motor cortex evoked a signal wave of around 4.69 ms latency and 4.13 mV amplitude in the hindlimb (Figure 2B and C). Following T8 SCI, the latency remained comparable but the amplitude of the signal was almost abolished (0.01 and 0.023 mV for left and right hindlimb, respectively, Figure 2B and C). Transplantation of iNSCs with biomaterial, but not biomaterial alone, significantly increased the amplitude to 1.42 ± 0.19 mV (around 35% that in naïve rats) at 7 months post-transplantation (Figure 2B and C). These results suggest that engraftment of iNSCs with biomaterial had functionally improved the cerebrospinal circuitry.

Figure 2.

Transplantation of iNSCs with biomaterial promotes electrophysiological recovery and locomotor improvement in SCI rats 7 months post-injury.

(A) Basso, Beattie and Bresnahan (BBB) scores in each SCI group from day 0 until 7 months post-SCI as examined. n = 6/group. (B) Latency and amplitude of motor evoked potentials (MEPs). n = 6/group. (C) MEPs were recorded in the left and right legs of rats in groups for different treatments and SCI only. Data are presented as mean ± SD. #P < 0.05, ##P < 0.01, vs. SCI + material alone group; *P < 0.05, **P < 0.01, vs. SCI only group (two-way analysis of variance with Sidak’s post hoc test). Normal, rats without SCI; SCI only, SCI rats that received vehicle; SCI + material alone, SCI rats that received biomaterial only; SCI + iNSCs transplantation, SCI rats that received iNSCs and biomaterial. n = 6/group. iNSCs: Induced neural stem cells; SCI: spinal cord injury.

Transplantation of iNSCs and biomaterial reduces lesion volume

To investigate the mechanisms underlying the behavioral and electrophysiological improvements in the engraftment groups, we performed pathological analysis of the spinal cord at 7 months after transplantation. The length and volume of the spinal lesion in each group were measured. The average lesion length was 4.1, 3.2, and 2.5 mm in the SCI only, SCI + material alone, and SCI + iNSCs transplantation groups, respectively (Figure 3A). Activated astrocytes (which are GFAP-positive) can be used to mark the border of spinal cord lesion areas. Using GFAP staining, the lesion volume in each group was calculated. Engraftment of iNSCs with material markedly reduced the lesion volume, which might have contributed to the functional recovery (Figure 3B). In addition, H&E staining was performed to examine tissue morphology (Figure 3C). Following SCI, the spinal cord exhibited a porous morphology with reduced cellular density (Figure 3C). This was reversed by transplantation with iNSCs-material (Figure 3C). Similarly, LFB staining to examine myelination, indicated a larger area of myelination at the lesion site in the iNSCs-material group compared with SCI only group (Figure 3D). These results suggest that engraftment of iNSCs with material can reduce the lesion volume and improve axon myelination, which might account for the functional recovery.

Figure 3.

SCI+iNSCs transplantation group shows reduced lesion volume compared with SCI only group at 7 months post-injury.

(A) Representative images of the spinal cord in each group. Red dotted boxes indicate lesion sites. (B) Lesion volumes of the injured spinal cord in each group. n = 3/group. (C) Hematoxylin and eosin staining of the spinal cord tissue in each group. The left panel shows a low magnification view of a horizontal section of the whole spinal cord (scale bar: 5000 μm). The right panel shows a higher magnification view of the black box inset in the left panel (scale bar: 50 μm). (D) Luxol fast blue staining of spinal cord tissue. The left panel shows a low magnification view of a horizontal section of the whole spinal cord (scale bar: 5000 μm). The right panel shows a higher magnification view of the black box inset in the left panel (scale bar: 50 μm). Data are presented as mean ± SD. *P < 0.05, **P < 0.01, vs. SCI only group (one-way analysis of variance with Tukey’s post hoc test). Normal, rats without SCI; SCI only, SCI rats that received vehicle; SCI + material alone, SCI rats that received biomaterial only; SCI+iNSCs transplantation, SCI rats that received iNSCs and biomaterial. iNSCs: Induced neural stem cells; SCI: spinal cord injury.

Engraftment of iNSCs with biomaterial promotes Tuj1 and NF200 expression in the lesion core

Transplanted iNSCs showed good survival on day 5 but this was gradually diminished over time. From day 15 onward, no transplanted iNSCs were detected in the spinal cord. To investigate what might account for the long-term functional benefits at 7 months post-transplantation, we examined the kinetics of the biomaterial used in the current study. Fibrinogen-specific antibody detects fibrin and can be used to examine kinetics of the biomaterial (Figure 4A and Additional Figure 4 ^{(715.1KB, tif)} ). Specifically, at 5 days post-injury/transplantation, a large amount of fibrin remained at the engraftment site (Figure 4A and Additional Figure 3 ^{(426.1KB, tif)} ). However, the biomaterial gradually degraded, with a significant decrease in quantity at 15 dpi. From 30 dpi onward, only a small amount of material was occasionally detected (Figure 4A). We also examined the presence of neuronal axons at the lesion core. Using GFAP staining to define the lesion borders, we stained spinal cord tissue sections for Tuj1 (a marker for immature neurons at the early stage of neuronal differentiation) and for NF200 (a marker for mature neurons). Compared with the SCI only group, some Tuj1 signal (Figure 4B) and NF200 signal (Figure 4C) (0.02–0.03%) was observed at the core of the lesion at lower magnification. At higher magnification (Figure 4D), only a few Tuj1⁺ or Tuj1⁺/NF200⁺ cells were detected at the lesion core in the SCI only group. At the core and border of the lesion center, a greater abundance of Tuj1⁺ and Tuj1⁺/NF200⁺ signals were detected in the iNSCs-material group (Figure 4D–F). The material alone group also showed a greater abundance of Tuj1⁺ and Tuj1⁺/NF200⁺ signals at the lesion core, in comparison with the SCI only group. Furthermore, by staining for MAP2 (a marker of mature neurons), and synapsin (a marker of presynaptic terminals), we detected synapsin⁺/MAP2⁺ possibly innervated axons at the lesion core in the iNSCs-material group, but not in the SCI only group (Figure 4G). These results indicate that engraftment of iNSCs and biomaterial could have led to the generation of functional axons at the lesion site.

Figure 4.

Degradation of fibrin, Tuj1, and NF200 expression at the lesion core and around the lesion border in each group 7 months after SCI.

(A) Staining and analysis of fibrinogen at different time points post-injury. A small amount of fibrin was detected from day 30 until 7 months post-SCI. Blue, 4′,6-diamidino-2-phenylindole (DAPI); green, human fibrinogen; Red, glial fibrillary acidic protein (GFAP). **P < 0.01, vs. 5 dpi. (B) Neuronal class III β-Tubulin (Tuj1) and GFAP staining 7 months post-injury. A higher amount of Tuj1 expression was detected at the lesion site of the SCI+iNSCs transplantation group. Blue, DAPI; Green, Tuj1; Red, GFAP. (C) Neurofilament-200 (NF200) and GFAP staining 7 months post-injury. A higher amount of NF200 expression was detected at the lesion site of the SCI+iNSCs transplantation group. Blue, DAPI; green, NF200; green; red, GFAP. (D) Tuj1 and NF200 expression at the lesion core and around the lesion border in each group. Blue, DAPI; green, Tuj1; red, GFAP; white, NF200. (E) Proportion of Tuj1⁺ cells at the lesion core and around the lesion border. n = 6/group. (F) Proportion of Tuj1⁺/NF200⁺ cells at the lesion core and around the lesion border. n = 6/group. (G) Staining of synapses at the lesion site. Neurons with synaptic connections were detected at the lesion core in the SCI+iNSCs transplantation group. Blue, DAPI; green, synapsin; red, microtubule-associated protein 2 (MAP2) indicating mature neurons. Left, low magnification. Right, high magnification of the white dotted box in the left image. Scale bars in A, D and the left image in G, 100 μm. Scale bars in B and C, 2000 μm. Scale bar in the right image of G, 25 μm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01 (one-way analysis of variance with Tukey’s post hoc test). iNSCs: Induced neural stem cells; SCI: spinal cord injury.

Transplantation of iNSCs and biomaterial has an impact on the microenvironment of the lesion site

Transplanted iNSCs showed good survival at 5 dpi, but this was gradually diminished over time. No transplanted iNSCs were detected from day 15 onward. Nevertheless, the effect on motor function and pathology appears to be long-lasting. One possibility was that during the acute and/or subacute stages following SCI, the hostile microenvironment might have abolished the key step(s) involved in endogenous axonal regeneration and/or neurogenesis. Modulation of the niche at the early stage might have a long-lasting effect on pathology and function. To address this question, we examined niche components of spinal cord tissue sections at 15 dpi. Staining for the trophic factors, brain-derived neurotrophic factor (BDNF), insulin-like growth factor 1, and neurotrophin-3, showed that only BDNF was detected, but with no significant difference observed between the three groups (Figure 5A and B). Staining for the cytokines, IL4, IL6, tumor necrosis factor-α (TNFα), and transforming growth factor-β (TGFβ) revealed that TGFβ expression was increased while TNFα expression was reduced in the iNSCs-material group (Figure 5A and B). Neither IL4⁺ nor IL6⁺ cells were detected at the lesion site at this time point. In addition, staining for the corresponding receptors, TrkB (receptor of BDNF), TGFβR1 (receptor of TGFβ), and TNFR1 (receptor of TNFα) showed lower expression of TGFβR1 and TrkB in the lesion core after SCI (Additional Figure 5A ^{(870KB, tif)} ). TGFβR1 expression was elevated and TNFR1 markedly reduced after transplantation of iNSCs (Additional Figure 5B ^{(870KB, tif)} ). No statistical significance was detected for TrkB staining among the three groups. These results are consistent with the previous staining results (Figure 5). Further results showed that the number of CD206⁺/Iba1⁺ cells (M2 microglia) was greater in the iNSCs-material group on 15 dpi, compared with the SCI only and SCI+material alone groups (Figure 6A and C). Staining of CD86 (a marker of M1 macrophages) showed that CD86⁺ cells were reduced at 15 dpi and 30 dpi (Additional Figure 6A ^{(999.7KB, tif)} ). On 5 dpi and 15 dpi, the quantity of CD86⁺ cells decreased following fibrin and/or iNSCs transplantation (Additional Figure 6B ^{(999.7KB, tif)} ). On 30 dpi, there was no significant difference among the three groups (Additional Figure 6B ^{(999.7KB, tif)} ). We believe that this might be due to the suppression of inflammatory responses and transition of macrophages/microglia to a M0 state at 30 dpi. We also examined immune cells by staining for CD45 and CD68, finding a reduced number of both CD45⁺ and CD68⁺ cells in the iNSCs-material group (Figure 6B and D). Previous studies have shown that the extracellular molecule, laminin, stimulated the growth of neurites from dorsal root ganglion neurons (Plantman et al., 2008; Anderson et al., 2016), therefore we examined laminin expression in each group. Staining for laminin and GFAP (Figure 7) revealed that the grafts did not alter the subtypes of activated astrocytes at the lesion site. These results indicate that iNSCs improved the microenvironment post-SCI, which might be beneficial for regeneration.

Figure 5.

BDNF, TGFβ, and TNFα expression in control, material, and SCI+iNSCs transplantation groups.

(A) Brain-derived neurotrophic factor (BDNF), transforming growth factor-β (TGFβ), and tumor necrosis factor-α (TNFα) staining in each group at 15 days post-injury (dpi). SCI + iNSCs transplantation group showed higher levels of TGFβ and lower levels of TNFα expression at the lesion site. Blue, DAPI; green, BDNF, TGFβ, and TNFα; red, glial fibrillary acidic protein (GFAP); white, ionized calcium-binding adaptor molecule 1 (Iba1). (B) Proportion of BDNF-, TGFβ- and TNFα-positive cells in each group at 15 dpi. n = 3/group. Scale bars, 100 μm. *P < 0.05, **P < 0.01 (one-way analysis of variance with Tukey’s post hoc test). iNSCs: Induced neural stem cells; SCI: spinal cord injury.

Figure 6.

SCI + iNSCs transplantation group shows reduced inflammation response after SCI.

(A) CD206 expression patterns in each group. Blue, 4′,6-diamidino-2-phenylindole (DAPI); green, CD206; red, glial fibrillary acidic protein (GFAP). Scale bars: 100 μm. (B) CD45 and CD68 expression patterns in each group at 15 days post-injury (dpi). Blue, DAPI; green, CD45 and CD68; red, GFAP. Scale bars, 100 μm. (C) The proportion of CD206-positive cells. n = 3 per group. *P < 0.05, **P < 0.01 (two-way analysis of variance with Sidak’s post hoc test). (D) The proportion of CD45- and CD68-positive cells. n = 3 per group. *P < 0.05, **P < 0.01 (one-way analysis of variance with Tukey’s post hoc test). iNSCs: Induced neural stem cells; SCI: spinal cord injury.

Figure 7.

Engraftment of iNSCs with material does not change the type of scar tissue 7 months post-injury.

(A) Whole section staining of glial fibrillary acidic protein (GFAP)/laminin. n = 6/group. White dotted boxes indicate the lesion region. (B) Regions in the white boxes of A. Blue, 4′,6-diamidino-2-phenylindole (DAPI); green, laminin; red, GFAP. Scale bars: 2000 μm. iNSCs: Induced neural stem cells; SCI: spinal cord injury.

Discussion

In the current study, we transplanted biomaterial (fibrin) alone or biomaterial plus human iNSCs into a rat transection SCI model, and found that engraftment of biomaterial alone could improve motor function and pathology in SCI rats. Addition of iNSCs to the biomaterial further enhanced motor function and pathology at the injured spinal cord in our rat model.

In our study, the biomaterial and human iNSCs only persisted for a short time post transplantation (up to about 1 week). Nevertheless, the impact appeared long-lasting. Seven months following the injury/transplantation, improvement in motor function and pathology were still observed. The reasons underlying this phenomenon are not fully understood, but could be attributed to changes in the microenvironment of the lesion site and the subsequent effect on pathological cascades following the primary traumatic event. Following the initial SCI, the damage continues to evolve and can be divided into several phases: acute (0–48 hours), subacute (2–14 days), intermediate (2 weeks to 6 months), and chronic (beyond 6 months) (Rowland et al., 2008). The different phases consist of correlated cascades, with one event leading to another. In particular, inflammation starts during the acute phase, with activation of glial cells and infiltration of immune cells following the disruption of blood vessels and the swelling and compression of cord tissue (Fan et al., 2018). During the subacute phase, the neurons and glia die, and the dead cells and released substances further bolster the inflammatory response and the invasion of immune cells, causing further damage to the spinal cord. This reinforcing feed forward loop strengthens itself and reaches a peak during the subacute phase, gradually tapering off during the intermediate/chronic phase. It is possible that the extent and magnitude of these earlier inflammatory cascades may determine the level of subsequent long-term functional damage. In turn, intervention at these earlier phases may subdue the amplifying cascades and exert long-lasting effects.

Furthermore, our findings suggest that iNSCs and/or the biomaterial might have suppressed the inflammatory cascades during the acute and/or subacute stages following SCI. This early intervention might have disrupted the feed forward loop, resulting in long-lasting pathological and behavioral effects. NSCs can produce neurotrophic factors and other soluble cytokines/chemokines (Hawryluk et al., 2012), which might have facilitated alterations of the microenvironment at the lesion site towards one that is conducive to the regenerative process. Microglia are the resident immune cells of the central nervous system and can be broadly classified into M1 and M2 phenotypes; the M2 phenotype is generally considered to be beneficial for central nervous system repair (Cherry et al., 2014; Wang et al., 2015). In contrast, chronic activation of M1 type microglia is part of the inflammatory response following SCI and can trigger further loss of neural tissues (Cherry et al., 2014; Wang et al., 2015). In the current study, engraftment of iNSCs and biomaterial was associated with a bias towards a M2 microglial phenotype, a reduction of proinflammatory cytokine TNFα, an increase of anti-inflammatory cytokine TGFβ, and fewer infiltrating immune cells into the lesion site.

In our study, iNSCs and/or the biomaterial survived for a short time in vivo, however they led to a long-lasting, beneficial effect. Previous studies have reported sub-optimal survival of NSCs transplanted into the injured spinal cord (Johnson et al., 2010; Du et al., 2011; Zou et al., 2020). This might be attributed to a shortage of neurotrophic factors in the harsh microenvironment of the injured spinal cord. The type of grafted cells may also be related to the survival rate. For example, Zou et al. (2020) found that human spinal cord-derived neural stem/progenitor cells (NSPCs) exhibited better survival rates and greater proliferation capacity than brain-derived NSPCs when transplanted into SCI rats with complete transection. To overcome this problem, researchers have tried different approaches to boost the survival of NSCs following transplantation, such as applying biomaterials and numerous growth factors. For example, Lu et al. (2012) transplanted the human fetal spinal cord-derived cell line (566RSC) embedded in growth factor-containing fibrin into athymic nude rats. They found that the grafts exhibited good growth and led to functional recovery. Robinson and Lu (2017) optimized the trophic support of NSCs embedded in fibrin matrices, which were implanted two weeks after a C5 lateral hemisection injury, resulting in good survival and neuronal differentiation of the graft. Without trophic support, the graft exhibited less viability, demonstrating the importance of growth factors. Rosenzweig et al. (2018) grafted human spinal cord-derived NPCs with a growth factor cocktail and fibrin–thrombin into the injured cervical spinal cord of rhesus monkeys under triple drug immunosuppression. The grafts survived at least 9 months post-injury.

The reasons for the better graft survival in the above studies could be due to: (1) physical and/or trophic support of NSC grafts from biomaterials and/or beneficial growth factors; (2) aspects of immune recognition. Grafts of lesser immune disparity were used in some studies: some studies employed allogeneic grafts, some used immunodeficient mice, and some applied human grafts into nonhuman primate recipients. In our study, human cells were transplanted into murine recipients; and (3) a different timing scheme of transplantation (for example, 2 weeks post-injury) when the inflammatory reaction has receded (Robinson and Lu, 2017).

The eventual clearance of iNSCs and biomaterial in SCI rats might result from the xenogeneic nature of the transplantation used in the current study. iNSCs can be reprogrammed from a patients’ own somatic cells and thus may be used as an autologous donor cell source. How autologous iNSCs might react to the niche of injured spinal cord (such as survival, migration, proliferation, and differentiation) remains an interesting question. We may gain some hints by first transplanting human iNSCs into an immunodeficient rat SCI model in future work.

This study has some limitations that should be noted. First, only female rats exhibiting less aggressive behavior than injured male rats were used as they are easier to care for post-injury. However, it is possible that sex-specific treatment effects may exist and future studies using both female and male rats are warranted. Second, we performed transplantation on the same day of injury, and the unfavorable inflammatory microenvironment may not be conducive to the survival of transplanted cells. The appropriate transplantation time might be 14 dpi when the inflammatory reaction has receded. Finally, the xenogeneic nature of human iNSCs used in our study would instigate immune system recognition in rats, despite the administration of immunosuppressive drugs (cyclosporine D in this study). Cyclosporine D might affect the SCI microenvironment and has been shown to reduce the proportion of CD4⁺ helper T cells (Fee et al., 2003) and inhibit the expression of IL-2 and IL-2 receptor. Cyclosporine D also suppresses the expression of TNF-α in SCI rats (Chen et al., 2018), and reduces the pool of the macrophage population (Setkowicz et al., 2009) and infiltration of macrophages in contused spinal cord (Ritfeld et al., 2010). Researchers transplanted MSCs combined with cyclosporine into a mouse model of skin injury and found that CD206+ immunosuppressive M2 subpopulations were significantly increased, while the frequency of CD45⁺CD11b⁺ cells were markedly reduced, indicating a beneficial role of cyclosporine (Hajkova et al., 2017). In addition, immune recognition of xenogeneic grafts and the anti-inflammatory effect associated with iNSCs and/or iNSC-derived soluble factors appear to be two opposite forces, therefore the anti-inflammatory effect of iNSCs might have played a major role. Further work is needed to address the detailed mechanisms underlying this anti-inflammatory effect.

In conclusion, in this study, we transplanted iNSCs reprogrammed from human PBMCs and/or thrombin plus fibrinogen into an injured spinal cord. Engraftment with iNSCs and biomaterial facilitated the recovery of motor and electrophysiological functions of SCI rats. Modulation of the microenvironment exerted by the iNSCs/biomaterial grafts may play crucial roles in the functional recovery after SCI. Nevertheless, a larger sample size and more behavioral tests would be helpful and necessary to confirm the efficacy for future clinical translation. In the current study, around 20 rats were employed in each group. In future efforts, more animals for each condition and more behavioral tests (such as BBB, electrophysiology detection, gait analysis, inclined plane test) should be used to consolidate the results of the current study.

Additional files:

Additional Figure 1 ^{(574.5KB, tif)} : Study flowchart.

Additional Figure 1

Study flowchart.

BBB test: Basso-Beattie-Bresnahan test; BDNF: Brain derived neurotrophic factor; dpi: days post injury; HE: hematoxylin-eosin; iNSCs: Induced neural stem cells; LFB: Luxol fast blue; MEP: motor evoked potential; SCI: spinal cord injury; TGFβ: Transforming growth factor-β; TNFα: tumor necrosis factor-α.

Additional Figure 2 ^{(289KB, tif)} : Experimental timeline.

Additional Figure 2

Experimental timeline.

BBB: Basso, Beattie and Bresnahan test; MEP: motor evoked potentials; SCI: spinal cord injury.

Additional Figure 3 ^{(426.1KB, tif)} : Staining of human fibrinogen on a whole spinal cord section of spinal cord injury animals sacrificed on day 5 post-transplantation.

Additional Figure 3

Staining of human fibrinogen on a whole spinal cord section of spinal cord injury animals sacrificed on day 5 post-transplantation.

Blue, DAPI; Green, fibrinogen; Red, glial fibrillary acidic protein (GFAP). Scale bar, 100 μm. n = 3 per group. DAPI: 4′,6-Diamidino-2-phenylindole.

Additional Figure 4 ^{(715.1KB, tif)} : Recognition of fibrin by anti-fibrinogen antibody.

Additional Figure 4

Recognition of fibrin by anti-fibrinogen antibody.

Fibrinogen was converted to fibrin by mixing with thrombin for different periods of time in vitro. The mixture was then stained with anti-fibrinogen antibody. A stable staining signal was observed at 5, 30 minutes, and 1 hour, indicating that anti-fibrinogen antibody detects both fibrinogen and fibrin. Green, human fibrinogen. Upper panels, immunofluorescence; lower panels, bright field images. Scale bars: 250 μm.

Additional Figure 5 ^{(870KB, tif)} : Receptors of BDNF, TGFβ, and TNFα staining in SCI only, SCI + material alone, and SCI + iNSCs transplantation groups.

Additional Figure 5

Receptors of BDNF, TGFβ, and TNFα staining in SCI only, SCI + material alone, and SCI + iNSCs transplantation groups.

(A) Staining for receptors of brain-derived neurotrophic factor (TrkB), transforming growth factor-β, (TGFβR1), and tumor necrosis factor-α (TNFαR1) in each group at 15 days post-injury (dpi). SCI+iNSCs transplantation group showed higher levels of TGFβR1 and lower levels of TNFαR1 expression at the lesion site. Blue, DAPI; Green, TrkB, TGFβR1, and TNFαR1; Red, glial fibrillary acidic protein (GFAP); White, ionized calcium-binding adaptor molecule 1 (Iba1). Scale bars, 100 μm. (B) Proportion of TrkB-, TGFβR- and TNFαR1-positive cells in each group at 15 dpi. n = 3 per group. *P < 0.05, **P < 0.01 (one-way analysis of variance with Tukey’s post hoc test). iNSCs: Induced neural stem cells; SCI: spinal cord injury.

Additional Figure 6 ^{(999.7KB, tif)} : CD86 expression patterns in each group.

Additional Figure 6

CD86 expression patterns in each group.

(A) CD86 staining in each group. Blue, 4′,6-diamidino-2-phenylindole (DAPI); green, CD86; red, glial fibrillary acidic protein (GFAP); White, ionized calcium-binding adaptor molecule 1 (Iba1). Scale bars, 100 μm. (B) The proportion of CD86-positive cells. n = 3 per group. *P < 0.05, **P < 0.01, vs. iNSCs+Material (two-way analysis of variance with Sidak’s post hoc test).

Additional Table 1: Antibodies used in immunofluorescence staining.