Activation of endogenous neurogenesis and angiogenesis by basic fibroblast growth factor-chitosan gel in an adult rat model of ischemic stroke

Abstract

Attempts have been made to use cell transplantation and biomaterials to promote cell proliferation, differentiation, migration, and survival, as well as angiogenesis, in the context of brain injury. However, whether bioactive materials can repair the damage caused by ischemic stroke by activating endogenous neurogenesis and angiogenesis is still unknown. In this study, we applied chitosan gel loaded with basic fibroblast growth factor to the stroke cavity 7 days after ischemic stroke in rats. The gel slowly released basic fibroblast growth factor, which improved the local microenvironment, activated endogenous neural stem/progenitor cells, and recruited these cells to migrate toward the penumbra and stroke cavity and subsequently differentiate into neurons, while enhancing angiogenesis in the penumbra and stroke cavity and ultimately leading to partial functional recovery. This study revealed the mechanism by which bioactive materials repair ischemic strokes, thus providing a new strategy for the clinical application of bioactive materials in the treatment of ischemic stroke.

Introduction

Stroke is a major cause of mortality, second only to ischemic heart disease (Lo et al., 2003; Pandian et al., 2018; Zerna et al., 2018). Ischemic stroke stimulates a series of complex pathological processes including inflammatory responses (Burda and Sofroniew, 2014), cellular necrosis or apoptosis, glial cell activation, and massive neuronal death in the stroke cavity and penumbra, ultimately resulting in functional loss in the corresponding brain areas (Kitamura et al., 2004; Burda and Sofroniew, 2014; Chovsepian et al., 2022).

Common clinical treatments for ischemic stroke include rapid thrombolytic therapy using recombinant tissue plasminogen activator (rtPA) and mechanical thrombectomy via surgical techniques within the golden time (within 3 hours of stroke). Nearly 90% patients receive conservative treatment instead of rtPA or thrombectomy because they miss the window for emergency intervention (Zhao and Willing, 2018). Current research into alternative treatments for stroke mainly focuses on two categories. The first is cell transplant therapy, which involves transplantation of different types of stem cells (for example, mesenchymal stem cells) that differentiate to replace the damaged cells and provide metabolic support for other endogenous cells in the stroke cavity; however, stem cell survival in the stroke cavity and integration with the endogenous tissue are poor (Dabrowski et al., 2019). In addition, the complexity of obtaining stem cells (in terms of sources, ownerships, and other ethical issues) also reduces their clinical application in this context. The second treatment category under investigation is the use of mini pumps to constantly deliver neurotrophic factors (such as vascular endothelial growth factor and basic fibroblast growth factor [bFGF]) into the stroke cavity or penumbra to protect damaged cells. However, these soluble factors have an extremely short half-life that limits their ability to persist in the stroke cavity or cross the blood-brain barrier, resulting in a low clinical efficacy.

Ischemic stroke can also spontaneously activate endogenous stem/progenitor cells to some degree, which represents a stress response to the ischemic process as well as potential self-repair. As previously reported (Kaneko et al., 2017), during ischemic stroke, some newborn neuroblasts (doublecortin (DCX)⁺ cells) may migrate a long distance from the subventricular zones (SVZ) along the vessels and astrocytes towards the stroke cavity and the surrounding areas, such as the cortex and corpus striatum. However, these cells cannot enter the stroke cavity to replace the damaged brain tissue, and their migration efficiency remains low, because of the joint impact of increasingly severe inflammatory responses in the stroke cavity, decreased blood supply, and hindrance by glial scar (Kaneko et al., 2017). The discovery that, after stroke occurs in adult mammals, endogenous stem/progenitor cells can be activated to produce newborn neurons that later migrate toward the stroke penumbra has attracted considerable interest in the research and clinical communities (Lindvall and Kokaia, 2015). The area around the stroke cavity (i.e., the penumbra) is usually defined as all tissue within 200–400 μm of the stroke cavity border. This area has been shown to exhibit robust neuronal and vascular plasticity (Carmichael, 2006; Brown et al., 2008). Delivering bioactive gel containing neurotrophic factors to the stroke cavity can induce vascular and axonal regeneration in the penumbra and facilitate the functional restoration to some extent (Nih et al., 2018; Xu et al., 2019). In both the photochemical occlusion model and the mild/moderate/severe cerebral artery occlusion model (Rust et al., 2019), various interventions, including pumping brain derived neurotrophic factor (BDNF) into the stroke cavity, intraperitoneally injecting cilostazol, providing an enriched environment, feeding a retinoic acid-containing diet, or delivering functional electric stimulation, increase the numbers of proliferative cells (5-bromo-2′-deoxyuridine (BrdU)⁺ cells), migrating cells (double positive for BrdU and DCX), and newborn neurons (BrdU⁺/neuronal nuclei (NeuN)⁺) in the subventricular zone (SVZ) and penumbra to some degree (Plane et al., 2008; Keiner et al., 2009; Tanaka et al., 2010; Diederich et al., 2012; Liu et al., 2013); however, neither BrdU⁺/NeuN⁺ nor BrdU⁺/DCX⁺ cells were ever observed in the stroke cavity.

Ischemic stroke causes damage directly by inducing tissue necrosis and indirectly by eliminating extracellular matrix that would otherwise physically support cell infiltration or tissue regeneration and promoting the formation of glial scar that blocks axon regeneration (Kaneko et al., 2018; Ferrari et al., 2022; Hernández and Pérez-Álvarez, 2022). The cavity is therefore a potential transplant location because it can accept a certain volume of implanted cells without affecting normal brain function (Nih et al., 2018). During the subacute period of ischemic stroke (3–14 days after stroke), rapid and complex pathological changes occur within and around (penumbra) the stroke cavity, including neuronal necrosis and apoptosis, inflammation, hypoxia/ischemia, and formation and stabilization of glial scars, which have a substantial impact on later interventions. The stroke cavity is a fibrotic region that is devoid of neurons and has a sparse, disordered vasculature. The cavity represents the area of brain tissue that is lost after stroke and is associated with the functional disability seen in patients post-stroke (Stokowska et al., 2017). We previously showed that, 7 days after stroke, almost all neurons within the stroke cavity had become necrotic, most of the blood vessels had disintegrated, microglial activation and glial scar density were continuing to increase, and the number of activated neural stem/progenitor cells in the SVZ had peaked, thus forming a stable cavity at the center of the area affected by stroke. We therefore chose the first 7 days after ischemic stroke as the optimal time window for intervention (Liang et al., 2020; Li et al., 2022).

As chitosan has an active role in cell proliferation, morphogenesis, and wound healing (Mo et al., 2010), it is often used to mediate long-term delivery of neurotrophic factors to the central nervous system. Ultimately, chitosan promotes neural stem cell survival and differentiation into neurons (Duan et al., 2016). It is unknown whether bioactive materials can repair the damage caused by stroke by activating endogenous neurogenesis and angiogenesis. This study aimed to explore endogenous neurogenesis and angiogenesis in an adult rat model of ischemic stroke after filling the stroke cavity with bFGF-loaded chitosan gel.

Methods

Preparation of bFGF-chitosan gel and bFGF release dynamics test

Under sterile conditions, chitosan (Sigma, St. Louis, MO, USA) was dissolved at a mass fraction of 3% in a 1% acetic acid (Sigma), and the remaining acetic acid was eluted. Genipin (TargetMol, Boston, MA, USA) was dissolved in deionized water to prepare the 8 mM solution. Next, the 3% chitosan solution and 8 mM genipin solution were mixed at a 1:1 ratio. After cooling to 4°C, bFGF (Sigma) was added to the mixture and slowly stirred for 72 hours at 4°C to obtain the gelatinous scaffold material, which was then stored in a refrigerator at 4°C. Chitosan-genipin solution without bFGF was prepared in a similar manner for use as a control.

As described previously (Duan et al., 2016), the chitosan gel prepared as described above with or without bFGF was added to the growth medium of brain P3 neural stem cells from four newborn (< 24 hours old) Wistar rats. For the control group, 100 ng/mL bFGF (Thermo Fisher Scientific, Waltham, MA, USA) was added to the growth medium daily. Supernatant samples were collected at 1, 3, 6, 12, and 1–9 weeks. ELISA was performed for each group to assess the bFGF release dynamics using an Emax immunoassay system (Promega, Madison, WI, USA) according to the manufacturer’s instruction. The ELISA plate (Promega) was coated with mouse anti-human bFGF polyclonal antibody (pAB). A 50 μL diluted sample was added to each well and incubated at 37°C for 1 hour. The liquid was removed from each well, without washing, and 100 μL of a biotin-conjugated antibody (1×) was added to each well and incubated for 1 hour at 37°C. Each well was then aspirated and washed three times. Next, 100 μL HRP-avidin (1×) was added to each well and incubated for 1 hour at 37°C. Then each well was aspirated and washed five times. Subsequently, 90 μL of the chromogenic substrate 3,3,5,5-tetramethylbenzidine (Sigma) was added to each well and incubated for 15–30 minutes at 37°C. Finally, 50 μL of stop solution was added to each well. An ELISA analyzer (BMG LABTECH Company, Offenburg, Germany) was used to read absorbance at 450 nm within 5 minutes of adding the stop solution. The bFGF content was determined via comparison to a previously generated bFGF standard curve.

Attenuated Total Reflection-Fourier Transform Infrared spectroscopy

Chitosan gel with or without bFGF was analyzed using a FTIR-7600 spectrometer with attenuated total reflectance (Lambda, Sydney, Australia). The Attenuated Total Reflection-Fourier Transform Infrared (ATR-FTIR) spectra were recorded at a resolution of 2 cm^–1 for 20 scans, from 4000 cm^–1 to 600 cm^–1.

Thermogravimetric analysis

Thermogravimetric analysis (TGA) of samples (about 10 mg each) was carried out with a Differential Scanning Calorimeter 1 (DSC 1, Mettler Toledo, Zurich, Switzerland). Chitosan gel with or without (control) bFGF was analyzed. TGA was performed at 25°C (room temperature) to 800°C, and the real-time TGA and differential thermogravimetric (DTG) values were recorded.

Animals

Seventy-two specific pathogen-free (SPF)-grade treatment-naïve male Wistar rats, weighing 250–300 g, were used in this study. The rats were maintained three to a cage at room temperature with 60% relative humidity and a 12/12-hour dark/light cycles. The rats were provided by the Experimental Animal Center of Capital Medical University (license No. SYXK (Jing) 2018-0003). All experiments were approved by the Experimental Animal Welfare Committee of Capital Medical University (approval No. AEEI-2019-140) on October 18, 2019, and were designed and reported according to the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines (Percie du Sert et al., 2020). Animals were randomly divided into three groups with 24 animals in each group: stroke only (stroke; only photothrombotic occlusion), lesion control (lesion; after photothrombotic occlusion, only stroke cavity suction, without any intervention), and treatment (after photothrombotic occlusion, injection of bFGF-chitosan gel into the stroke cavity).

Preparation of the ischemic stroke model via photothrombotic occlusion and treatment

The rats were anesthetized by intraperitoneal injection with 1% pentobarbital sodium (25 mg/kg), kept supine, and sterilized with iodine. An incision was made on the skin of the inner thigh, and the subcutaneous tissue was separated to expose the femoral vein. Next, the rats were fixed on a stereotaxic apparatus, the skin on the top of the head was sterilized with iodine and incised longitudinally to expose the skull, and the connective tissue was removed from the skull surface. A window was made in the skull bone 2 mm forward of the anterior fontanelle and 3 mm left lateral of the midline (Paxinos and Watson, 1998) while keeping the dura mater intact. Normal saline was dripped onto the dura mater until the superficial brain vessels could be seen clearly. Then, a 532-nm fixed-wavelength laser (GPD105-M-12, Taiwan Shuanzhou Photoelectric Co., Ltd., Taiwan, China) was used to illuminate the bone window, with the laser diameter set at 2 mm and the distance to the brain surface fixed at 1 cm. At the same time, Rhodamine (80 mg/kg body weight, concentration 40 mg/mL, Sigma) was injected into the femoral vein through the previously made incision (during preparation of the optical embolization model, rose red was injected, which formed an embolism after irradiation through the bone window). After injection, the needle was pulled out and sterile cotton balls were used to stop the bleeding. Illumination was delivered for 10 minutes, causing cortical infarct in the rat motor sensory area (Rust et al., 2019b). Then the thigh and head muscles were sutured, and the skin was sterilized with iodine. In the sham group, a skull window was opened, but no other procedures were performed.

Seven days after stroke, different interventions were performed. Under sterile conditions, the skull window was re-exposed. Then, an incision was made in the dura mater and an in-house biological tissue suction device (ZL 2019 1 1140256.1) was used to suction 0.2 mm³ of tissue from the stroke cavity, followed immediately by filling the stroke cavity with 2 mg bFGF-chitosan gel (Hao et al., 2017). Because a previous study showed that chitosan gel alone does not promote functional recovery after stroke (Liang, 2020), a chitosan-only control was not used in this study. Next, the dura mater, muscles, and skin were sutured, and the wound was sterilized. After the operation, the animals were replaced in their cages. Penicillin (2 units/100 g, North China Pharmaceutical, Shijiazhuang, China) was injected for the following 3 days to prevent inflammation.

Histological sampling

At specified time points after stroke or intervention, the rats were euthanized with 1% pentobarbital sodium (30 mg/kg, MilliporeSigma) injected intraperitoneally. After heart perfusion with 20 mL normal saline followed by 40 mL 4% paraformaldehyde (Aladdin, Shanghai, China), the brain was carefully removed, placed at 4°C, fixed with 4% paraformaldehyde for 1 day, and dehydrated for 1 day in 30% sugar phosphate buffer (PB). Next, the dehydrated brain tissue was placed in a –20°C cryostat for 30 minutes, then embedded in Optimal Cutting Temperature (O.C.T.) compound (Sakura Tokyo, Japan), and sectioned into 30-μm continuous coronal slices using a cryostat microtome (CM1850 cryostat microtome, Leica, Wetzlar, Germany). Six sets of complete brain slices were taken from each rat, and each set was immunohistochemically stained. To track the fate of proliferative cells (Plane et al., 2008), after the tissue was suctioned from the stroke cavity, rats received intraperitoneal injection of BrdU (50 mg/kg body weight, in 0.9% NaCl solution, Sigma) twice daily for 7 days to observe the effect of injury and repair on cell proliferation.

Immunohistochemical staining

The slices prepared at different time points were immunohistochemically stained. The primary antibodies used are listed in Table 1. Each slice was rinsed with 0.01 M PBS solution for three times for 5 minutes each and then incubated with 10% normal goat serum (Zhongshan Golden Bridge, Beijing, China) at room temperature for 60 minutes, followed by incubation with the primary antibody at 4°C overnight. Then, the slices were rinsed again and incubated with an appropriate secondary antibody (Table 1) at room temperature for 7 hours. Nuclei were labeled with 4′,6-diamidino-2-phenylindole (DAPI); Vector Laboratories, Inc. Burlingame, CA, USA) for 10 min. Finally, the slices were sealed with PB glycerol (Sigma) for imaging. The slices were observed, and cells were counted manually, using a laser confocal microscope (SP8 laser confocal scanning microscope, Leica). ImageJ V1.8.0.112 (National Institutes of Health, Bethesda, MD, USA) was used to performed regional fluorescence intensity analysis and length measurement (Schneider et al., 2012). To identify double-labeled cells, double/triple-staining images were collected in sequential mode using a confocal microscope (TCS SP8, Leica). Ten 30-μm-thick sections spaced 180 mm apart spanning the injured area were harvested. Each BrdU-positive cell was examined under high magnification in its full z dimension, and only those cells whose BrdU-positive nucleus was unambiguously associated with a given marker (Nestin, DCX, Tuj1, NeuN, Glut-1) were considered double-labeled. The number of cells expressing various markers within a counting frame (25 μm × 25 μm) was determined. Data are expressed as the average number of double-labeled cells divided by the average number of single-labeled cells. The vascular density was determined by calculating the glut-1 fluorescence per unit area. Data are expressed as the average glut-1 fluorescence signal area divided by the area of the entire visual field.

Table 1.

Primary and secondary antibodies used in immunohistochemical staining

Antibody	Animal species	Dilution	Cat#	Supplier
Doublecortin (a marker for neuroblasts)	Rabbit	1:200	ab18723	Abcam, Cambridge, UK
BrdU (a marker for proliferative cells)	Mouse	1:200	ZM-0013	ZSGB Biotechnology, Beijing, China
Nestin (a marker for neural stem cells)	Rabbit	1:100	ab93157	Abcam
Tuj1 (a marker for immature neurons)	Rabbit	1:500	T200020	Sigma, St. Louis, MO, USA
NeuN (a marker for mature neurons)	Rabbit	1:500	ab177487	Abcam
Glut-1 (a marker for vascular endothelial cells)	Rabbit	1:200	ab115730	Abcam
PDGFR-β (a marker for pericytes)	Rabbit	1:200	SAB4502148	Sigma
Alexa Fluor 488 anti-mouse IgG	Goat	1:200	A-11001	Molecular Probes, Carlsbad, CA, USA
Alexa Fluor 488 anti-rabbit IgG	Goat	1:200	A-11008	Molecular Probes
Alexa Fluor 594 anti-rabbit IgG	Goat	1:200	A-11012	Molecular Probes
Alexa Fluor 647 anti-chicken IgY	Goat	1:200	A-21449	Molecular Probes

Evaluation of sensory and motor function after ischemic stroke

For the cylinder experiment (limb asymmetry experiment), the rats were placed in a transparent round glass cylinder, and the number of times each forelimb touched the wall (left front, right front, and bilateral) out of 20 total touches was calculated. Each rat was evaluated three times (Knoth et al., 2010). This test was performed to detect the decline in forelimb motor function in rats with ischemic stroke caused by photochemical embolism.

For the grid experiment (foot fault experiment), the rats were placed on an overhead grid with a grid size of 2.3 cm × 2.3 cm. The number of hind limb false steps (three toes of the hind limb or all of the hind limb falling below the grid) and normal steps within 20 steps was counted. Each rat was evaluated three times. This test was performed to detect the sensorimotor decline in the hindlimb after ischemic stroke caused by photochemical embolism.

Statistical analysis

No statistical methods were used to predetermine sample sizes; however, our sample sizes were similar to those reported in a previous publication (Hao et al., 2017). Unless otherwise specified, all data are denoted as mean ± SEM. The Shapiro-Wilk normality test was used to assess normal distribution of data. Levene’s test was used to assess homogeneity of variance. One-way analysis of variance was performed to compare multiple groups, and the least significant difference post hoc test was performed to analyze differences between two groups. All statistical analyses were performed using GraphPad Prism 7.0 software (GraphPad Software, San Diego, CA, USA, www.graphpad.com). P < 0.05 was considered to be statistically significant.

Results

Dynamics of controlled bFGF release by bFGF-chitosan gel

In the group of cells that was supplemented with soluble bFGF daily, no bFGF was detected in the supernatant at any time point, which indicates that bFGF has a very short half-life at 37°C. In the group that was treated with chitosan without bFGF, no bFGF was detected at any time point. In the bFGF-chitosan gel group, bFGF was detected in the supernatant for at least 9 weeks (Figure 1). This findings suggest that bFGF-chitosan gel released bFGF stably and constantly from 1 hour to 9 weeks.

Figure 1.

In vitro basic fibroblast growth factor (bFGF) elease from the bFGF-chitosan gel over 9 weeks.

Data are shown as the mean ± SD, n = 7/group. bFGF: Basic fibroblast growth factor; w: week.

bFGF was successfully grafted onto chitosan gel

To explore whether bFGF was successfully grafted onto the chitosan gel, bFGF, chitosan, and the bFGF-chitosan gel were analyzed by FTIR. The results showed that, in the bFGF-chitosan gel spectrum, the bending vibration absorption peak of amino N-H at 1593 cm^–1 seen in soluble bFGF and chitosan alone disappeared, while the absorption peak of the amide II band and the rocking vibration absorption peak of the amino group in the amide bond appeared at 1537 cm^–1 and 1151 cm^–1, respectively (Figure 2). This may be because the amino groups in the chitosan reacted with bFGF, indicating that bFGF was successfully grafted onto the chitosan gel.

Figure 2.

FTIR transmission spectra of the bFGF-chitosan gel, chitosan, and bFGF.

Compared with bFGF and chitosan, the bFGF-chitosan gel does not exhibit a bending vibration absorption peak at 1593 cm^–1, while it does exhibit an absorption peak at 1537 cm^–1 and a rocking vibration absorption peak at 1151 cm^–1. n = 6/group. bFGF: Basic fibroblast growth factor; FTIR: Fourier Transform Infrared spectroscopy.

Thermogravimetric analysis

To further investigate the thermal stability of the bFGF-chitosan gel, we analyzed the TGA curves of bFGF, chitosan, and bFGF-chitosan gel and obtained the DTG curves. These curves showed that, as the temperature increased from room temperature to 800°C, bFGF experienced one weight loss, which peaked at a decomposition temperature of 283°C (Figure 3A). In comparison, chitosan experienced two weight losses, which peaked at decomposition temperatures of 59°C and 295°C, because of water loss and decomposition of the chitosan, respectively. Finally, the bFGF-chitosan gel experienced three weight losses, which peaked at decomposition temperatures of 46°C, 149°C, and 279°C. Compared with soluble bFGF and chitosan alone, all decomposition temperatures for the bFGF-chitosan gel were shifted towards the left (Figure 3B), suggesting slightly lower thermal stability and indicating that bFGF was successfully grafted onto chitosan. The TGA, DTG, and FTIR results for the bFGF-chitosan gel suggest that bFGF interacts with chitosan through amino groups. This interaction, likely due to hydrogen bonding, might slightly reduce the thermal stability of the bFGF-chitosan gel.

Figure 3.

TGA and DTG curves for the bFGF-chitosan gel, chitosan, and bFGF.

(A) bFGF exhibited one weight loss, which peaked at a decomposition temperature of 283°C; chitosan exhibited two weight losses, which peaked at decomposition temperatures of 59 and 295°C; the bFGF-chitosan gel exhibited three weight losses, which peaked at decomposition temperatures of 46, 149, and 279°C. (B) Compared with soluble bFGF and chitosan alone, all decomposition temperatures for the bFGF-chitosan gel were shifted towards the left. n = 3/group. bFGF: Basic fibroblast growth factor; DTG: differential thermogravimetric; TGA: thermogravimetric analysis.

Application of bFGF-chitosan gel after stroke activates neural stem/progenitor cells

Previous studies showed that, 7 days after stroke in a rat model, a large amount of proliferated neuroblasts (i.e., DCX⁺/BrdU⁺ cells) are present in the SVZ, the glial scar around the stroke cavity is not fully mature, and the microglia-mediated inflammatory response has peaked, which promotes survival of migrated neural progenitor cells (Lian, 2020; Liang et al., 2020; Li et al., 2022). We therefore chose 7 days after stroke as the optimal time point for intervention.

Seven days after the stroke cavity was filled with bFGF-chitosan gel, a significantly greater number of nestin-positive cells was observed in the stroke cavity compared with the stroke only sand lesion control (LC) group (P < 0.05; Figure 4A, B, and E).

Figure 4.

The bFGF-chitosan gel activated neural stem/progenitor cells.

(A) Overview of the experimental timeline. BrdU was injected intraperitoneally immediately after the second operation and then twice daily for 7 consecutive days. (B) Nestin (red) and BrdU (green) double-positive cells were detected in the stroke cavity 7 days after the second operation for all groups. The images labeled i-I “, II-II “, and III-III “ show progressively higher-magnification images of the same regions, and z-stacks of the areas indicated with white boxes are shown on the right. (C, D) Migrating DCX (red) and BrdU (green) double-positive cells were detected in the SVZ region of the bFGF-chitosan gel group at 7 and 14 days after the second operation, while no double-positive cells were detected in the stroke only and lesion control groups. The images labeled i-I “, II-II “, and III-III “ show progressively higher-magnification images of the same regions, and z-stacks of the areas indicated with white boxes are shown on the right. (E, F) Quantification of the immunostaining results shown in B–D (n = 6 rats/group, *P < 0.05, vs. stroke alone group; #P < 0.05, vs. lesion control group, one-way analysis of variance with the least significant difference post hoc tests). *Undegraded chitosan. Data are presented as mean ± SE. bFGF: Basic fibroblast growth factor; BrdU: bromodeoxyuridine; DAPI: 4′,6-diamidino-2-phenylindole; DCX: doublecortin.

No DCX⁺ cells were observed in the stroke cavity in the stroke only (14 days after stroke) or LC (7 days after aspiration). In comparison, 7 days after bFGF-chitosan gel application, more DCX⁺/BrdU⁺ cells were observed in the SVZ, and some even migrated to the area around the stroke cavity (P < 0.05; Figure 4C and F). At 14 days after aspiration, the neurogenesis provoked by stroke had almost ceased in the stroke only and LC groups, whereas in the bFGF-chitosan gel group, many DCX⁺/BrdU⁺ cells were observed in the SVZ, and more such cells had migrated towards the stroke cavity (P < 0.05; Figure 4D and F).

bFGF-chitosan gel facilitates the generation and long-term survival of neurons in and around the stroke cavity

After ischemic stroke, bFGF-chitosan gel strongly activated DCX⁺/BrdU⁺ cells migrate to the damaged area. However, whether these activated cells could differentiate into neurons remained unclear. At 28 days after bFGF-chitosan gel application, many Tuj1⁺/BrdU⁺ immature neurons were observed in and around the stroke cavity, while almost no Tuj1⁺/BrdU⁺ cells were observed in the stroke cavity in the stroke only and LC groups (P < 0.05; Figure 5A and B). Three months after bFGF-chitosan gel application, NeuN⁺/BrdU⁺ cells were observed in the stroke cavity, but none were observed in the stroke cavity of the stroke only and LC groups (Figure 5C and D). These results suggest that bFGF-chitosan gel facilitates neuronal generation in the stroke cavity in the short term and promotes the survival and maturation of these neurons in the long term.

Figure 5.

Generation and maturation of neurons in the stroke cavity after application of the bFGF-chitosan gel.

(A) Immunofluorescence staining for Tuj1 (red) and BrdU (green) revealed immature newborn neurons (Tuj1⁺ /BrdU⁺ cells) in the stroke cavity in the bFGF-chitosan gel group at 1 month after the second operation, while no immature neurons were detected in the stroke only and lesion control groups. Nuclei were labeled with DAPI (blue), and serially magnified images are shown. Z-stacks of confocal images demonstrated co-labeling of cells with BrdU and Tuj1 in the stroke cavity in the bFGF-chitosan gel group. (B) Quantification of the immunostaining results shown in A (n = 6, *P < 0.05, vs. stroke only group; #P < 0.05, vs. lesion control group, one-way analysis of variance, the least significant difference test ). (C) NeuN⁺ (red) and BrdU⁺ (green) cells were detected in the stroke cavity of the bFGF-chitosan gel group at 3 months after the second operation. (D) Quantification of the immunostaining results shown in A (n = 6 rats/group, *P < 0.05, vs. stroke alone group; #P < 0.05, vs. lesion control group, one-way analysis of variance with the least significant difference post hoc tests). The dotted boxes indicate the enlarged areas. *Undegraded chitosan. Data are presented as mean ± SE. bFGF: Basic fibroblast growth factor; BrdU: bromodeoxyuridine; DAPI: 4′,6-diamidino-2-phenylindole; Tuj1: β-tubulin III.

bFGF-chitosan gel facilitates angiogenesis in and around the stroke cavity

Angiogenesis is a critical event during neural regeneration after ischemic stroke. After stroke, reconstruction of the local micro-circulating environment may prolong the survival of relevant tissues including neurons and ultimately improve the efficacy of rehabilitation of stroke patients (Slevin et al., 2006). Nih et al. (2018) reported that neuronal survival after ischemic stroke was correlated with increased blood supply. To investigate whether bFGF-chitosan gel facilitates angiogenesis after photothrombotic occlusion, we used Glut-1 to label vascular endothelial cells. In the stroke only group, at 7–14 days after stroke, the stroke cavity was empty and devoid of blood supply, with almost no blood vessels remaining. In comparison, at 7 days after bFGF-chitosan gel application in the bFGF-chitosan gel group, many newborn vascular endothelial cells (Glut-1⁺/BrdU⁺ cells) were observed in the stroke cavity, together with an greater percentage of vascular area compared with the stroke only and LC groups (P < 0.05, Figure 6C). At 14 days after bFGF-chitosan gel application, we observed similar results (Figure 6A and D). These findings suggest that bFGF-chitosan gel facilitates the long-term regeneration of vascular endothelial cells in the stroke cavity.

Figure 6.

The bFGF-chitosan gel facilitates angiogenesis and blood vessel maturation in and around the stroke cavity.

(A) Immunofluorescence staining for Glut-1 (red) and BrdU (green) revealed a large number of Glut-1⁺ and BrdU⁺ newborn vessels in and around the stroke cavity in the bFGF-chitosan gel group, while little neovascularization was observed in the stroke only and lesion control groups. (B) Immunofluorescence staining for Glut-1 (red), BrdU (green), and Pdgfr-β (marker of perivascular cells, gray) was performed 3 months after the second operation. Serially magnified images are shown. Z-stack of confocal images demonstrate co-labeling of cells with BrdU and Glut-1 in the stroke cavity. (C, D) Quantification of Glut-1 fluorescent area and the number of Glut-1⁺/BrdU⁺ cells in the stroke cavity at 7 and 14 days after the second operation. n = 6 rats/group, *P < 0.001, vs. stroke alone group; #P < 0.001, vs. lesion control group (one-way analysis of variance with the least significant difference post hoc tests). *Undegraded chitosan. The dotted boxes indicate the enlarged areas. Data are presented as mean ± SEM. bFGF: Basic fibroblast growth factor; BrdU: bromodeoxyuridine; DAPI: 4′,6-diamidino-2-phenylindole; Pdgfr-β: platelet-derived growth factor receptor beta.

To investigate whether these vascular endothelial cells could develop into functional vascular networks, we assessed vascular maturity 1 and 3 months after bFGF-chitosan gel application. During angiogenesis, pericyte coverage is considered an indicator of vascular maturity and function (Zhang et al., 2012). Thus, we used platelet-derived growth factor receptor beta (Pdgfr-β) to label pericytes to assess vascular maturity (Zhang et al., 2012). Three months after bFGF-chitosan gel application, Glut-1/BrdU/Pdgfr-β–triple positive mature, functional vessels were observed in the stroke cavity (Figure 6B), suggesting that bFGF-chitosan gel could not only facilitates vascular endothelial cell generation in the short term, but also promotes the maturation and function of these vessels in the long term.

bFGF-chitosan gel facilitates functional restoration after stroke

Damage to different brain regions leads to different changes in behavioral functions; therefore, preclinical studies usually use corresponding behavioral tests to evaluate functional recovery related to different brain regions. The most typical tests include the cylinder test and grid foot fault test, as they reflect forelimb dexterity and hindlimb sensory and motor deficits in rodents (Cheng et al., 2014; Tennant et al., 2017; Caracciolo et al., 2018), as well as the effects of different interventions on sensory and motor function after stroke. After stroke, deficits in forelimb and hindlimb function were observed in all groups. One day after stroke, the number of wall placements of the ipsilateral forelimb of all groups significantly decreased (P < 0.05, Figure 7A), while the foot fault number of the ipsilateral hindlimb sharply increased (P < 0.05, Figure 7B). There were no statistically significant differences in forelimb and hindlimb functional deficits among the groups.

Figure 7.

The bFGF-chitosan gel promoted functional recovery after stroke.

(A) Right-paw touch test results for the three groups over 12 weeks. At 8 weeks after the second operation and thereafter, the bFGF-chitosan gel group exhibited a significantly higher rate of right-paw touches compared with the other groups. (B) Foot fault test (grid test) results for the three groups over 12 weeks. At 4 weeks after the second operation and thereafter, the rate of foot faults in the bFGF-chitosan gel group was significantly reduced compared with that in the other groups (mean ± SEM, n = 6 rats/group, *P < 0.05, vs. other two groups [one-way analysis of variance with the least significant difference post hoc tests]). bFGF: Basic fibroblast growth factor.

At 4–12 weeks after application of the bFGF-chitosan gel, rats in the stroke only and LC groups showed some spontaneous recovery of forelimb and hindlimb function, although it was very slow and restricted to a very low functional level. Significant sensory and motor function deficits persisted until week 12 (Figure 7A and B).

Next we asked whether the bFGF-chitosan gel, which induced neurogenesis and angiogenesis after stroke, could also improve functional deficits. To test this, we assessed rat forelimb and hindlimb behavior in the different groups at various time points. The results showed that, at 4 weeks after stroke, rats in the bFGF-chitosan gel group exhibited significant improvement in the grid test, but relatively slow improvement in the cylinder test, compared with the stroke only and LC groups (P < 0.05; Figure 7A and B). These improvements lasted until 12 weeks after stroke (P < 0.05; Figure 7A and B). These results suggest that the bFGF-chitosan gel sustainably and significantly improved sensory and motor function deficits after photothrombotic occlusion.

Discussion

To investigate whether bioactive materials can promote endogenous neurogenesis to repair the damage caused by stroke, we filled the stroke cavity with bFGF-containing chitosan gel and intraperitoneally injected rats with BrdU daily for 1 week. Morphological and behavioral evaluations were then performed to investigate whether bFGF-containing chitosan gel promoted the proliferation, migration, and differentiation of endogenous neural stem cells into neurons, promoted the formation of a functional vascular network, and improved motor function. The results showed that the bFGF-chitosan gel strongly stimulated vascular network formation within and around the stroke cavity, induced neurogenesis within and around the stroke cavity, and promoted partial recovery of the behavioral functions.

In this study, we prepared a chitosan gel that mediates long-term release of bFGF. The chitosan gel has three main advantages that may contribute to its successful function. First, it contains genipin as the crosslinking agent, which has less cytotoxicity and better biocompatibility than other crosslinking agents (Sung et al., 1999; Mi et al., 2020). Second, we strictly controlled the temperature while generated the gel, unlike Yang et al. (2011): to maximize the activity of the neurotrophic factor, the entire process was performed at 4°C. Finally, the gel form makes this treatment suitable for a stroke model, as it is able to slowly release bFGF into the damaged area. The bioactive gel slowly released bFGF over 9 weeks at 37°C, leading to long-term activation of neural stem/progenitor (Nestin⁺/BrdU⁺, DCX⁺/BrdU⁺) cells, recruitment of these cells to the stroke cavity and its surrounding areas, where they differentiated into neurons, maintenance of the long-term survival of these newborn neurons, and facilitation of vascular regeneration and maturation, which ultimately led to some degree of functional recovery.

In previous studies (Zhang et al., 2021, 2022), a dual site-selective functionalized (DSSF) poly (β-amino esters) strategy was developed using bio-orthogonal chemistry to promote nerve regeneration in the brain. bFGF-loaded nanohybrid hydrogel, vascular endothelial growth factor–coated heparin nanoparticles, or BDNF-containing hydrogel was applied to the stroke cavity to facilitate regeneration of vessels and NF⁺ axons in the penumbra, which resulted in partial functional restoration in a rat model of stroke (Zhang et al., 2012; Cook et al., 2017; Jian et al., 2018); however, no newborn vessels, neurons, or axons were observed in the stroke cavity (Cook et al., 2017; Nih et al., 2018). Other studies have delivered a large amount of vascular endothelial growth factor, FGF-2, and EGF to the lateral ventricle via a micro-pump and found that this approach protects neurons to some extent, reduces neuronal apoptosis, and stimulates the generation of newborn neurons in the brain. However, these protocols require strict dose control to avoid blood-brain barrier leakage and brain edema (Ma et al., 2001; Nakatomi et al., 2002; Kaya et al., 2005). In our previous study, we found that combining bFGF-containing sodium hyaluronate gel application and exogenous neural stem cell transplantation repaired the damage caused by traumatic brain injury in adult rats (Duan et al., 2016). The transplanted biomaterial improved the hostile microenvironment and facilitated differentiation of exogenous neural stem cells into neurons that ultimately formed synaptic connection with the host brain (Duan et al., 2016). Nevertheless, sodium hyaluronate gel degrades quickly, resulting in low efficiency of bFGF packaging and release (Duan et al., 2016). In the present study, we designed a chitosan gel that mediates controlled release of bFGF over a long period of time (9 weeks). Compared with previously reported bFGF-loaded chitosan materials, the bFGF-chitosan gel generated in this study can be injected, and can better fill the damaged area and promote damage repair. Our gel activated endogenous neural stem/progenitor cells in the SVZ over a long time period, recruited these cells to the stroke cavity and its surrounding areas, where they subsequently formed new mature neurons (NeuN⁺/BrdU⁺), and promoted the long-term survival of these new neurons (3 months after gel application). However, it remains unclear whether these newborn neurons can reconstruct the cerebral cortex and integrate into the host neural circuit. In the future we will address these issues via neurotropic virus tracing, photo-genetics in combination with the patch clamp technique, or in vivo electrobiological tests.

Brain vessels play a critical role in maintaining brain structure and functions, and normal blood supply is a prerequisite for brain functions A previous study (Rust et al., 2019) reported that angiogenesis and axonal regeneration are closely correlated with improvement in functional disability after stroke. Rust et al. (2019) reported that using endostatin to inhibit angiogenesis significantly reduced axonal regeneration and functional recovery. This indicates that restoration of blood supply has a significant effect on functional recovery in later stages after stroke.

Generation of vascular endothelial cells in the stroke cavity was greatly promoted by bFGF-chitosan gel application and was accompanied by an increase in vessel density. Three months after transplantation, newborn vessels coated with pericytes (Glut-1/BrdU/Pdgfr-β–triple-positive cells) were observed in the stroke cavity, suggesting that the bFGF-chitosan gel promoted formation of mature, functional blood vessels and restored bilateral hindlimb function as assessed by the the cylinder experiment and foot fault experiment.

Functional recovery is the ultimate goal of stroke treatment. Behavioral tests are used to evaluate changes in function in response to different interventions. The cylinder test has good sensitivity for evaluating deficits in spontaneous use of the rodent forelimb after stroke. When placed in a transparent glass cylinder, stroke rats will rear up on their hindlimbs to actively explore vertical surfaces with their forelimbs and whiskers, displaying an asymmetry in forelimb use (Davis et al., 1997). Furthermore, the grid test can be used to objectively evaluate sensory and motor deficits. When placed on an elevated, level grid, normal rats will grasp the metal frame precisely when moving their feet, while rats suffering from sensory and motor deficits will slip and fall when moving their ipsilateral limb.

As stroke destroys neural circuits in the brain, the brain responds to pathological breakdown by attempting to re-establish its intrinsic function, but the spontaneous recovery process is slow and insufficient. We observed in this study that, 1 day after photothrombotic occlusion, sensory and motor functions declined sharply in all rats. Over time, these deficits spontaneously improved to some degree, although 3 months after stroke the rats in the LC and stroke only groups still suffered from long-term disabilities.

The application of bFGF-chitosan gel significantly increased the number of ipsilateral forelimb wall placements, while decreasing hindlimb foot faults, demonstrating significant functional improvement. However, this study did not evaluate functional neural circuit reconstruction after stroke or explored the origin of the new neurons that appeared in the regenerated tissues. In the future, we will use a chemical genetics approach (such as the hM4Di-CNO system) to silence newborn neurons and neural circuits or use endostatin to inhibit newborn vessels to determine the correlation between the functional recovery and newborn neurons or vessels.

Conclusions

In this study, we demonstrated that, 7 days after ischemic stroke caused by photothrombotic occlusion, application of bFGF-chitosan gel to the stroke cavity effectively facilitated angiogenesis and stimulated neural stem/progenitor cells to proliferate, migrate, and differentiate into neurons. The gel also promoted survival and maturation of newborn neurons in the stroke cavity, further enhancing functional recovery. These findings may shed new light on the clinical treatment of stroke in the subacute (3 days to 7 months after stroke) and chronic (7 months after stroke) phases.