An integrated hydrogel-V3 interneuron therapy promotes functional repair of spinal cord injury via neural circuit reconstruction and microenvironment remodeling

Abstract

Spinal cord injury (SCI) disrupts neural circuits and creates an inhibitory microenvironment, posing challenges such as low cell survival rates and limited host integration for traditional cell transplantation therapies. This study developed an injectable, self-adaptive, and self-repairing oxidized hyaluronic acid-carboxymethyl chitosan (OHA-CMCS) dual-network interpenetrating hydrogel. This hydrogel serves as a functionalized, dynamically responsive cell-matrix co-delivery platform for delivering spinal cord-specific V3 neuronal precursors derived from human pluripotent stem cells. Through tissue-mimetic mechanical design, the hydrogel closely simulates the spinal cord tissue microenvironment. Its reversibly crosslinked network exhibits excellent compliance and self-healing capabilities, forming bidirectional feedback coupling with V3 cells across mechanical and biochemical dimensions, thereby significantly enhancing cell survival and functional maturation. In rats with complete spinal cord transection, the “material-cell synergistic system” (OC0.33+V3) formed by the OHA-CMCS hydrogel and V3 cells markedly improved motor function (BBB score, grip strength, gait analysis) and remodeled the injured microenvironment. Mechanistic studies reveal that this system drives microenvironmental reprogramming through material-cell interactions, inhibiting glial scar formation, inducing M2 polarization of microglia, and promoting axonal regeneration and vascular remodeling. Chemogenetic validation further confirms that transplanted V3 neurons successfully integrate into host neural circuits and exert inhibitory regulatory functions. This study proposes a dual-engine strategy of “material-driven regulation and cell-function integration,” revealing the mechanism by which biomimetic hydrogels synergize with neurons to repair spinal cord injury, establishing a new paradigm for intelligent neuroregeneration systems.

Graphical abstract

Highlights

• •Injectable self-healing hydrogel delivers V3 neural cells for spinal cord repair.
• •Biomimetic hydrogel enhances cell survival via bidirectional feedback coupling.
• •System improves motor function and remodels microenvironment in SCI models.
• •Mechanism inhibits glial scar and promotes axonal regeneration.
• •Transplanted neurons integrate into host circuits with functional validation.

1. Introduction

Spinal cord injury (SCI) is a highly debilitating neurological disorder caused by trauma, typically resulting in severe sensory and motor function loss below the level of injury, and even lifelong disability [1]. The World Health Organization (WHO) estimates that 250,000 to 500,000 new cases of SCI occur worldwide each year [2], with lifetime medical costs ranging from $1 million to $5 million per patient, placing a heavy economic and emotional burden on families and society [3]. Currently, there is still a lack of effective treatment options for SCI [4]. In recent years, cell replacement therapy has provided new insights into SCI repair [5]. Various types of cells, including Schwann cells [6], neural stem/progenitor cells [7,8], and mesenchymal stem cells [9], have been investigated for SCI treatment. Among them, neural progenitor cells (NPCs), due to their self-renewal ability and potential for multi-lineage differentiation, are considered highly promising transplantable cell types that can replace lost neurons in the injured area and promote neural circuit reconstruction [10]. However, traditional NPCs transplantation still faces several challenges, such as low survival rates, inefficient migration, and poor integration in the hostile injury microenvironment [11]; NPCs often primarily differentiate into astrocytes, which may exacerbate glial scar formation and even induce neuropathic pain [12,13].

To achieve more precise neural repair, researchers have proposed the strategy of transplanting precursor cells of specific neuronal subtypes. SCI causes the loss of specific types of motor neurons and interneurons, and transplanted neural progenitor cells can mature in vivo, integrate into host neural circuits, replace some lost functions, and ultimately promote motor function recovery [10,[14], [15], [16]]. Among various candidate cells, precursor cells of spinal V3 interneurons exhibit unique advantages. V3 interneurons are core elements of the spinal central pattern generators (CPGs) [17] and are crucial for generating and controlling rhythmic movements, such as walking, running, and swimming [[18], [19], [20]]. Therefore, supplementing V3 interneurons is an important strategy for restoring motor function after SCI.

The advent of human pluripotent stem cell (hPSC) technology has made it possible to obtain human ventral spinal p3/V3 precursor/progenitor cells (hereafter referred to as V3 progenitor cells) via directed differentiation in vitro, providing a potential source for cell transplantation. Despite the tremendous potential of V3 progenitor cells transplantation, low survival and retention rates in the inflammatory and inhibitory microenvironment after SCI remain major obstacles. Injectable biomaterials, especially hydrogels, are considered an effective solution [21]. An ideal hydrogel not only fills the injury cavity through minimally invasive methods but also provides physical protection for transplanted cells, isolates inhibitory signals from the host, and creates a favorable microenvironment for cell survival and integration [[22], [23], [24], [25]]. Natural polymer hydrogels are highly favored due to their excellent biocompatibility and bioactivity [[26], [27], [28]]. In this study, we selected hyaluronic acid (HA) and chitosan (CS) as the hydrogel matrix materials. HA is an important extracellular matrix component with the potential to regulate inflammation and reduce glial scar formation while exhibiting viscoelastic properties similar to natural spinal cord tissue [[29], [30], [31]]; CS has been shown to inhibit scar formation and promote axonal regeneration [32]. More importantly, both materials have been demonstrated to possess neuroprotective and regeneration-promoting activities: hyaluronic acid fragments can modulate the phenotype of microglia and suppress inflammatory storms; while chitosan derivatives neutralize negatively charged inhibitory molecules in the injury zone, clearing obstacles for axonal growth. By combining them in a specific ratio, we aim to construct a functional scaffold that not only mimics the mechanical properties of spinal cord tissue (viscoelasticity, toughness) but also continuously releases bioactive signals through material-cell dynamic interactions. This scaffold will provide V3 progenitor cells with full-cycle support from transplant survival to functional integration.

Based on this rationale, the present study designed and characterized an oxidized hyaluronic acid (OHA)-carboxymethyl chitosan (CMCS) composite hydrogel in terms of its rheological properties, self-healing behavior, and biocompatibility. It systematically evaluated the effects of this material system on the viability, proliferation, and differentiation lineage of V3 progenitor cells. Subsequently, using a nude rat complete spinal cord transection model, the study validated the ability of the material-cell composite system to remodel the injured microenvironment and restore motor function. We anticipate that this strategy, which deeply integrates material properties with cellular functional requirements, will overcome the survival limitations of traditional cell transplantation and achieve dual-path synergy between “material-guided microenvironment reconstruction” and “cell-driven neural circuit regeneration,” thereby establishing a new paradigm for spinal cord injury repair.

This schematic illustrates the construction of OHA-CMCS hydrogel for spinal cord repair and its synergistic mechanism with V3 cell therapy. The OHA-CMCS hydrogel is formed by crosslinking oxidized hyaluronic acid with carboxymethyl chitosan via a dynamic Schiff base reaction. Its mechanical properties match those of natural spinal cord tissue and effectively support the survival, differentiation, and maturation of human spinal V3 neural progenitor cells in three-dimensional in vitro culture. In a rat spinal cord injury model, the transplanted OHA-CMCS hydrogel and V3 cells formed a synergistic therapeutic system: this system provided structural support by filling the injury cavity, laying the foundation for repair. Simultaneously, it drove the repair process through multiple active mechanisms, including suppressing neuroinflammation and glial scar formation, promoting angiogenesis to improve the local microenvironment, and significantly enhancing the survival and integration efficiency of transplanted V3 cells. The synergistic action of these mechanisms ultimately succeeded in promoting axonal regeneration and myelin repair, significantly improving hindlimb motor function and bladder function in rats (see Fig. 1).

Fig. 1.

Schematic of injectable self-healing hydrogel combined with V3 neuron therapy for spinal cord injury.

2. Results

2.1. Preparation and characterization of OHA-CMCS hydrogel

The various properties of OHA-CMCS hydrogel were systematically analyzed and validated. First, the gelation process (Fig. S1A) shows that OC hydrogel exhibits good gelation ability at room temperature, with the liquid hydrogel quickly transforming into a stable solid gel. Scanning electron microscopy (SEM) images (Fig. 2A, Fig. S1B) reveal that OC0.33 hydrogel has a uniform porous structure, which facilitates its biocompatibility and material exchange properties. We verified the structures of the two modified polymers through ¹H NMR and FTIR spectral analysis. Spectral analysis of OHA (Fig. 2B and C) confirmed the successful introduction of aldehyde groups, with its oxidation degree calculated via ¹H NMR to be 17.7% [33,34]. For CMCS, characteristic absorption peaks at 1592 cm⁻¹ and 1411 cm⁻¹ in the FTIR spectrum (Fig. 2D and E), corresponding to the asymmetric and symmetric stretching vibrations of the carboxyl group, respectively, provided conclusive evidence for the introduction of the carboxyl group. Furthermore, the FTIR spectrum of the hydrogel (Fig. 2F) detected a characteristic absorption peak at 895 cm⁻¹ for the Schiff base bond (C=N), confirming the successful formation of the hydrogel via a Schiff base reaction between OHA and CMCS [35].

Fig. 2.

Characterization of OHA-CMCS hydrogel. (A) SEM image of OHA-CMCS hydrogel. Scale bar: 400 μm. (B) Infrared spectra of HA and OHA. (C) ¹H NMR spectra of HA and OHA. (D) Infrared spectra of CS and CMCS. (E) ¹H NMR spectra of CS and CMCS. (F) Infrared spectra of OHA-CMCS hydrogel. (G) Time-sweep rheological analysis of different hydrogels at a fixed frequency of 1 Hz and 1% strain. (H) Frequency-sweep rheological tests (0.1–10 rad/s) for different hydrogels. (I) Swelling ratios of different hydrogels in PBS (n = 3). (J) In vitro degradation profiles of different hydrogels (n = 3). (K) Strain amplitude sweep (0.1–1000%) of the OC hydrogel at a fixed frequency of 1 Hz. (L) Step-strain measurements performed on OC0.33 hydrogel by oscillating between low (1%) and high (600%) strain at a fixed frequency of 1 Hz. (M) Schematic illustration of the dynamic covalent bonds responsible for the self-healing mechanism. Data are presented as mean ± SD.

Gelation time is a critical parameter for hydrogels [36]. The gelation time measurement results (Supplementary Table S1) indicate that with increasing CMCS ratio, the gelation time decreases, and three hydrogel formulations (OC1, OC0.5, and OC0.33) were selected for subsequent studies. The water content results of the hydrogels (Fig. S1C) show that the average water content of all three groups exceeded 95%. This hydrophilicity provides a favorable microenvironment for nutrient diffusion and cell survival. The mechanical properties of the hydrogel were tailored to simulate the in vivo matrix, which regulates cell fate: first, the mechanical properties of the matrix are crucial for stem cell behavior [27,37]; furthermore, softer microenvironments have been shown to preferentially guide neural stem cells towards neuronal differentiation [38]. Rheological experiments (Fig. 2G and H) showed that both time and frequency sweep tests demonstrated that the G′ of all hydrogels was significantly higher than G″, indicating typical elastic solid behavior and good structural stability, which meets the requirements for three-dimensional cell culture scaffolds [39]. Notably, OC0.33 hydrogel exhibited a storage modulus of 103.8 ± 13.9 Pa, which is comparable to the modulus of natural spinal cord tissue (100–3000 Pa) [40]. Further shear rheology tests (Fig. S1D) showed that OC0.33 hydrogel has good shear-thinning properties, and its viscosity rapidly decreases with increasing shear rate, making it suitable for minimally invasive implantation into defective sites [41]. Swelling and degradation experiments demonstrated that the OC0.33 hydrogel achieved swelling equilibrium within 4 h (Fig. 2I), exhibiting potential for rapid absorption of exudate. By day 28, the OC0.33 hydrogel maintained 51.7% ± 3.4% residual mass (Fig. 2J), revealing a slow degradation profile capable of providing stable three-dimensional scaffolding for tissue regeneration. Thermogravimetric analysis (TGA) results indicate that all hydrogels exhibit good thermal stability within physiologically relevant temperature ranges (Fig. S1E).

The self-healing ability of the hydrogel is crucial for its long-term stability in vivo and for the recovery of structure and function after injection [39]. Macroscopic experiments showed that hydrogel blocks, stained with rhodamine B and methylene blue, could autonomously heal within 2 min after being cut and brought into contact, maintaining structural integrity under gravity (Fig. S1F and G). Rheological tests further revealed its dynamic mechanical behavior. When the strain reached 341%, the storage modulus (G′) and loss modulus (G″) intersected, indicating that the hydrogel structure was disrupted at this critical point (Fig. 2K). In the continuous step-strain test, when the strain increased abruptly from 1% to 600%, G′ rapidly decreased and fell below G″, causing the material to transition from gel to sol state; however, when the strain returned to 1%, G′ quickly recovered to the initial level, indicating that the structure could rapidly rebuild (Fig. 2L–Fig. S1H). This excellent self-healing and recovery property is primarily attributed to the reversible cross-linked network formed by dynamic Schiff base bonds, which can continuously undergo hydrolysis and reformation in an aqueous environment (Fig. 2M) [42,43]. In summary, OHA-CMCS hydrogel demonstrates rapid self-healing, injectability, and high structural recovery ability, making it suitable for minimally invasive injection into irregular spinal cord injury sites, maintaining long-term structural integrity, and showing great potential for applications in neural repair [44].

2.2. In vitro cell compatibility and in vivo biocompatibility evaluation of OHA-CMCS hydrogel

Excellent biocompatibility is a prerequisite for the application of hydrogels in SCI [45]. To systematically evaluate the biocompatibility of OHA-CMCS hydrogel, this study assessed its in vitro cell compatibility and in vivo systemic toxicity.

First, the toxicity of the hydrogel to SH-SY5Y cells was assessed using the CCK-8 assay and live/dead cell staining. The results showed (Fig. S2A) that under culture with hydrogel extracts of different volume ratios, cell viability remained above 90% at 24 h, 48 h, and 72 h, with no significant difference compared to the negative control group (p > 0.05). Specifically, the OC0.5 and OC0.33 groups exhibited cell viabilities of 91.1 ± 3.0% and 93.6 ± 1.8% at 72 h, respectively. These results suggest that OC hydrogel does not exhibit cytotoxicity. Live/dead staining images showed no obvious dead cell signals (Fig. S2B), and the quantitative results were consistent with the previous findings (Fig. S2C), further confirming the good in vitro cell compatibility of OC hydrogel.

Since the hydrogel will undergo degradation upon implantation, the biocompatibility of the hydrogel and its degradation products is crucial [46]. To assess this, OC0.33 hydrogel was implanted into the SCI injury site in rats and compared with the control group. After 16 weeks, major organs (heart, liver, spleen, lung, kidney) from the rats were collected and stained with H&E. The results showed that the morphology of organs from all experimental groups remained normal, with no obvious pathological changes, indicating that OC0.33 hydrogel did not cause systemic toxicity (Fig. S2D). These results indicate that OC hydrogel exhibits good biocompatibility and safety at both the cellular level and in the overall animal model, providing necessary evidence for its use as a cell carrier in spinal cord injury repair.

2.3. In vitro induction of human spinal V3 interneurons and the effect of OC0.33 hydrogel on their maturation

During spinal cord ventralization, the gradient of Sonic Hedgehog (SHH) concentration determines the fate of ventral neuronal precursor cells. The highest concentration of SHH specifically induces the formation of the p3 lineage, which ultimately differentiates into spinal V3 interneurons [47] (Fig. S3A). V3 interneurons, as glutamatergic excitatory interneurons, play a key role in coordinating rhythmic movements in spinal cord neural circuits. To obtain this specific neuronal subtype, this study optimized and adjusted previous spinal motor neuron differentiation protocolsand established a four-stage differentiation system for human embryonic stem cells (hESCs) to spinal V3 interneurons (Fig. S3B) [48,49].

In the first week, hESCs were exposed for 7 days to neural induction medium (50% DMEM/F12, 50% Neurobasal, 1×NEAA, N2) supplemented with 2 μM SB431542, 2 μM DMH1 and 3 μM CHIR99021 (Fig. S3B), yielded highly enriched SOX1⁺/HOXC4⁺ caudal neural epithelia (NEPs) that formed typical neural rosettes (Fig. S3C and D). In the second stage, 1 μM smoothened agonist (SAG), 2 μM IWR-1, 2.5 μM IWP-2, and 0.1 μM retinoic acid (RA) were added to pattern NEPs into ventral spinal progenitor cells (Fig. S3B). Immunofluorescence showed that the majority of cells expressed NKX2.2 (Fig. S3E and F), confirming the successful acquisition of spinal p3 characteristics. The third stage involved prolonging the incubation time with SHH agonists, cultivating the differentiated spinal progenitor cells into neurospheres in suspension culture (Fig. S3B). During this stage, cells highly expressed the spinal V3 interneuron-specific transcription factor SIM1 (Fig. S3G and H) [50]. In the fourth stage (D21-D42), cells further differentiated into mature neurons, expressing neurofilament (NF200) and glutamate transporter (VGLUT2) (Fig. S3I and J), confirming their glutamatergic excitatory neuron properties. These results demonstrate that we successfully established an efficient and stable differentiation system for human pluripotent stem cells into spinal V3 interneurons, obtaining a highly purified and phenotypically defined cell population. Based on this system, we further evaluated the effects of OC0.33 hydrogel, known for its excellent cell compatibility, on the survival, proliferation, and differentiation of human spinal V3 progenitor cells.

Morphological observation showed that after 1 day of culture, the human spinal V3 progenitor cells enriched in neurospheres from both the OC0.33 hydrogel group and the control group could dissociate well and differentiate into neurons. After 7 days of culture, the cells in both groups further matured into typical V3 interneurons, with no obvious morphological differences observed (Fig. 3A). To verify their neuronal subtype properties, we performed immunofluorescence staining for mature neuron markers MAP2 and V3-specific glutamatergic marker VGLUT2. Both groups of cells exhibited high expression of MAP2 and VGLUT2 and displayed typical neuronal morphology (Fig. 4B). Confocal Z-axis scanning and three-dimensional reconstruction showed that MAP2 and VGLUT2 signals were highly co-localized in the OC0.33 hydrogel group (Fig. 3C). Quantitative analysis (Fig. 3D–E) revealed no significant differences in the fluorescence intensity of MAP2 and VGLUT2 between the OC0.33 hydrogel group and the control group (p > 0.05), indicating that the OC0.33 hydrogel did not interfere with the normal establishment of the cytoskeleton or the specific maintenance of neurotransmitter phenotypes during the differentiation of V3 progenitor cells into mature functional neurons. In conclusion, this study not only successfully established an efficient and stable differentiation system for spinal V3 interneurons but also confirmed that OC0.33 hydrogel effectively supports the differentiation and maturation of this specific neuronal population. These findings provide crucial experimental evidence for the clinical application of this strategy in spinal cord tissue engineering.

Fig. 3.

The OC hydrogel promotes the differentiation and maturation of human spinal cord V3 neural progenitor cells. (A) Representative morphology of V3 intermediate neurons cultured in OC0.33 hydrogel versus control conditions at days 1, 3, 5, and 7. Scale bars: 100 μm (Day1), 50 μm (Day3-7). (B) Immunofluorescence images of cells stained for the mature neuronal marker MAP2 (red) and the V3-specific glutamatergic marker VGLUT2 (green). Scale bar: 50 μm. (C) Higher-magnification view of the boxed area in (B), showing detailed spatial colocalization of MAP2 and VGLUT2 within the cells. Scale bar: 20 μm. (D, E) Quantitative analysis of (D) MAP2 and (E) VGLUT2 fluorescence intensity (n = 3). Data are presented as mean ± standard deviation (SD). Statistical differences were determined using one-way ANOVA. ns, not significant (p > 0.05).

Fig. 4.

Combinatorial therapy with OC0.33 hydrogel and V3 neural progenitor cells promotes long-term motor functional recovery after SCI in rats. (A) Schematic illustration of the rat spinal cord injury model and treatment. (B) Survival rates of rats in each group over the 16-week post-operative period. (C) Dynamic changes in body weight. (D) Changes in BBB scores in rats. The number of rats in each group at the end of the experiment: SCI (n = 7), OC0.33 (n = 6), V3 (n = 9), and OC0.33+V3 (n = 9). (E) Distribution of final BBB scores across groups at week 16. (F) Quantitative analysis of hindlimb grip strength (n = 5). (G) Proportion of effective footfalls during the horizontal ladder test at week 16 (n = 5). (H) Representative images of hindlimb gait patterns at week 16. (I) Representative waveforms of CMAP recorded from different groups. (J, K) Quantitative analysis of (J) CMAP amplitude and (K) latency across groups (n = 5). All quantitative data are presented as mean ± SD. Statistical differences were determined using one-way ANOVA for comparisons among multiple groups at a single time point (G, J, K) and two-way ANOVA for repeated measures over time (D, F). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

2.4. OC0.33+V3 combined treatment significantly promotes long-term motor function recovery after SCI

To evaluate the long-term therapeutic effects of OC0.33 hydrogel, human spinal V3 neural progenitor cells (V3), and their combined application (OC0.33+V3) on spinal cord injury (SCI), a complete T9 spinal cord transection model was established in nude rats. The motor function recovery of the rats was systematically evaluated over a 16-week observation period. As shown in Fig. 4A, after exposing the spinal cord at the T9 level, the OC0.33 hydrogel precursor solution was mixed with V3 cells to form a composite gel, which was then implanted into the injury site. The experimental animals were divided into four groups: SCI group (injury only), OC0.33 group (injury + gel implantation), V3 group (injury + cell transplantation), and OC0.33+V3 group (injury + gel and cell combined transplantation). The survival rate in the OC0.33+V3 group remained at 100% throughout the 16-week observation period, significantly reducing early mortality after SCI and contributing to better long-term prognosis (Fig. 4B). In the first two weeks post-surgery, the body weight of all groups decreased slightly and then stabilized (Fig. 4C).

For long-term motor function recovery, we used the Basso, Beattie, and Bresnahan (BBB) scoring scale for continuous evaluation over 16 weeks. Immediately after surgery, the rats exhibited complete paralysis of both hind limbs (BBB score = 0), confirming successful model establishment. Notably, at the 16-week time point, the BBB score of the SCI group was 6.0 ± 0.76, with the OC0.33 group (6.6 ± 0.74) and V3 group (7.0 ± 0.62) showing some improvement, but recovery remained limited. In contrast, the OC0.33+V3 group had a BBB score of 7.8 ± 0.48 at 16 weeks, significantly higher than the SCI group, with some animals scoring 9 (Fig. 4D and E).

Additionally, the hindlimb grip strength test showed that at 16 weeks post-surgery, the hindlimb grip strength in the OC0.33+V3 group recovered to 321.0 ± 20.66 g, significantly higher than the SCI group (269.1 ± 12.14 g) (Fig. 4F). This result indicates that the combined treatment effectively maintained hindlimb muscle strength and control ability during long-term recovery.

For motor coordination, the horizontal ladder walking test was conducted to assess fine motor abilities. The OC0.33+V3 group performed significantly better than the SCI group, with a notably higher proportion of effective steps, reflecting superior motor coordination maintained during long-term recovery (Fig. 4G). In gait analysis, SCI rats consistently exhibited a typical “scraping” dragging motion of the hindlimbs, while the OC0.33+V3 treated rats demonstrated frequent, weight-bearing coordinated gait with clear joint flexion-extension (Fig. 4H and supplementary video).

Supplementary video related to this article can be found at https://doi.org/10.1016/j.bioactmat.2026.05.034

The following are the supplementary data related to this article:

To further evaluate the recovery of neural signal conduction, electrophysiological testing was performed at 16 weeks post-surgery. The results showed that, compared to the SCI group (7.1 ± 2.79 mV), the OC0.33 and V3 groups both showed some increase in the amplitude of compound muscle action potentials (CMAPs), and the OC0.33+V3 group exhibited a significantly higher CMAP amplitude (12.5 ± 1.21 mV) (Fig. 4I–J). Regarding latency, the OC0.33+V3 group (0.6 ± 0.12 ms) showed a significant reduction compared to the SCI group (1.3 ± 0.56 ms) (Fig. 4K). In conclusion, the combined application of OC0.33 hydrogel and human spinal V3 progenitor cells significantly promotes long-term functional recovery after spinal cord injury, demonstrating substantial therapeutic value and clinical potential for spinal repair.

2.5. OC0.33+V3 combined treatment promotes recovery of spinal cord structure and metabolic function

Functional recovery is closely related to the repair of neural structures and the pathophysiological environment of the injury [51]. We comprehensively assessed the effects of OC0.33+V3 cell therapy on injury site structural repair and metabolic function recovery through macroscopic morphological observation, PET-CT metabolic imaging, and tissue section staining.

First, at 16 weeks post-surgery, the spinal cord was collected for pathological analysis. As shown in Fig. S4A, the SCI group exhibited obvious tissue loss in the injury area, forming large cavities and dense scar tissue, indicating severe neural tissue loss and repair failure. In contrast, the OC0.33+V3 group showed better tissue continuity, with a significantly reduced cavity size and a tissue morphology closer to normal spinal cord.

PET-CT imaging of glucose metabolism in the injury site and adjacent tissues revealed that the SCI group exhibited significantly reduced 18F-FDG uptake in the injury center (indicated by the white arrow) and surrounding areas, suggesting severely impaired neural metabolic activity. In contrast, the OC0.33+V3 group showed a diffuse high signal with significantly higher 18F-FDG uptake compared to the SCI group (Fig. S4B), indicating enhanced neural metabolic activity, possibly related to synaptic integration of newly formed neurons and restoration of neural circuit function.

Further histological analysis using H&E and Masson staining of the injury site revealed that the SCI group exhibited large cavitary structures (Fig. S4C), with an average cavity area of 2.1 ± 0.65 mm², indicating severe tissue loss. In contrast, the OC0.33+V3 combined treatment group showed significant structural repair, with a marked reduction in cavity area to 0.3 ± 0.12 mm², 84.6% smaller than the SCI group (Fig. S4D). Additionally, a large number of cells infiltrated the area, suggesting an active tissue regeneration response.

Masson staining revealed that the SCI group exhibited large amounts of disordered blue collagen fiber deposition at the injury edge, indicating widespread fibrous scar formation. In contrast, the OC0.33+V3 group showed significantly reduced collagen deposition, with more ordered alignment (Fig. S4E). Quantitative analysis of collagen volume fraction (CVF) revealed that the CVF value in the OC0.33+V3 group (7.4 ± 1.58%) was significantly lower than that of the SCI group (17.1 ± 2.86%) (Fig. S4F). These results suggest that the combined hydrogel and cell intervention effectively regulated extracellular matrix remodeling and suppressed the fibrotic process. In conclusion, the OC0.33+V3 treatment not only promoted structural repair of the spinal cord injury but also improved local metabolic function and effectively suppressed fibrous scar formation, providing a favorable morphological foundation and microenvironmental support for neural regeneration.

2.6. OC0.33 hydrogel enhances survival and integration of transplanted V3 neurons in spinal cord injury

A core challenge in cell transplantation is the low survival rate of transplanted cells, with many cells losing viability post-transplantation. To assess the effect of OC0.33 hydrogel on improving cell survival, we performed STEM121 immunofluorescence staining to compare the distribution of human-derived cells in the V3 transplantation group versus the OC0.33+V3 combined treatment group.

The results showed that in the V3 transplantation group, almost no STEM121-positive signal was observed in the injury center, suggesting that, in the absence of physical support and biological protection, transplanted cells struggled to survive in the inhibitory environment. In contrast, in the OC0.33+V3 combined treatment group, although a few scattered STEM121-positive cells were observed in the injury center, a large number of clustered positive cells were distributed in the rostral and caudal regions (Fig. 5A), indicating that OC0.33 hydrogel effectively promoted the engraftment and retention of transplanted cells in the peripheral regions of the injury, providing regional protection. Further quantitative analysis confirmed these observations. The percentage of STEM121-positive area in the injury center of the OC0.33+V3 combined treatment group was 8.2%, approximately 4.9 times higher than that in the single V3 transplantation group (Fig. 5B).

Fig. 5.

Transplanted human V3 neurons survive and integrate into circuits in vivo to improve motor function. (A) Representative images of STEM121 (red) and NeuN (green) immunofluorescence staining in the V3 and OC0.33+V3 groups at 16 weeks post-SCI. Scale bars: 500 μm (overview), 20 μm (inset). (B) Quantitative analysis of STEM121+ positive area percentage in the lesion center and adjacent rostral/caudal regions (n = 3). (C) Quantitative analysis of NeuN + positive area percentage in the lesion center and adjacent rostral/caudal regions (n = 3). (D) Representative immunofluorescence images of NeuN (green) and SYP (red) co-staining in the peri-lesion area of each group at 16 weeks post-SCI. Scale bar: 50 μm. (E) Quantitative analysis of SYP + positive area percentage in the peri-lesion area (n = 3). (F) Quantitative analysis of hindlimb grip strength in SCI rats before and after administration of DCZ, SALB, and solvent at 16 weeks post-transplantation (n = 6). (G) Quantitative analysis of hindlimb effective step rate in SCI rats during the horizontal ladder test before and after administration of DCZ, SALB, and solvent at 16 weeks post-transplantation (n = 5). (H) Schematic of experimental design. Human embryonic stem cells co-expressing Bi-DREADDs (hM3Dq and KORD) were differentiated into spinal cord V3 neural progenitor cells and transplanted into transected spinal cords post-injury. DCZ, SALB, and solvent were sequentially administered, followed by behavioral testing. All quantitative data are presented as mean ± SD. Statistical differences were determined using one-way ANOVA for comparisons among multiple groups at a single time point (E) and two-way ANOVA for repeated measures over time (B, C, F, G). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

More importantly, among these surviving human-derived cells, we observed a high proportion co-expressing the mature neuron marker NeuN (STEM121+/NeuN+), indicating that the transplanted V3 progenitor cells successfully differentiated into mature neurons. In comparison, the NeuN-positive signal intensity and distribution density were lower in the single V3 transplantation group than in the combined treatment group (Fig. 5C).

To assess the differentiation and maturation of the transplanted cells as well as synaptic integration, we performed immunofluorescent co-staining for NeuN (a marker of mature neurons) and SYP (synaptophysin, a presynaptic marker) on spinal cord injury site sections 16 weeks post-transplantation, and conducted a semi-quantitative analysis of the SYP-positive area. The results showed that in the V3 cell-hydrogel composite transplant group, NeuN-positive cells were widely distributed in the perilesional area, and the cell density was significantly higher than in the other groups, consistent with the results of previous studies. SYP staining revealed a denser distribution of SYP-positive granules in the perilesional area of the composite transplant group. Semi-quantitative analysis indicated that the percentage of SYP-positive area (22.3 ± 3.83%) was significantly higher than that in the SCI control group (1.3 ± 0.89%) and the V3 group (11.9 ± 4.44%) (Fig. 5D and E). These results demonstrate that transplanted V3 neurons are capable of differentiating into mature neurons and receiving synaptic inputs, suggesting that they have integrated into the host spinal cord neural circuits.

To elucidate the direct contribution of transplanted human spinal V3 neurons to functional recovery, we employed chemogenetic (DREADD) technology for functional verification. We introduced a dual DREADD system expressing both excitatory (hM3Dq) and inhibitory (KORD) receptors into human V3 progenitor cells and transplanted them into the injury core area of the spinal cord transection rat model (Fig. 5H). Behavioral analysis showed that the activity level of transplanted neurons could specifically regulate the degree of motor function recovery. After activation of transplanted neurons with the hM3Dq ligand DCZ, the rats’ hindlimb grip strength (316.1 ± 36.13 g) and effective step rate in the horizontal ladder (25.6 ± 9.91%) remained at a high level. In contrast, after inhibition of transplanted neuron activity with the KORD ligand SalB, hindlimb grip strength (283.9 ± 22.85 g) and effective step rate (12.4 ± 4.15%) significantly decreased, significantly lower than the functional level after activation (Fig. 5F and G). These reversible functional suppression results directly prove that the transplanted human V3 neurons have successfully integrated into the host motor neural network and contributed specifically to motor function recovery after spinal cord injury. In summary, the OC0.33 hydrogel promotes the survival, differentiation, and synaptic integration of transplanted human V3 neurons in a rat model of spinal cord injury. The OC0.33+V3 combination strategy facilitated functional recovery by supporting neuronal engraftment and integration into host circuits. These results suggest that biomaterial-assisted cell transplantation is a viable strategy for repairing spinal cord injury.

2.7. OC0.33+V3 combined treatment promotes nerve regeneration and inhibits glial scar formation after spinal cord injury

To further investigate the potential mechanisms of the OC0.33+V3 combined treatment in spinal cord injury (SCI) repair, we systematically assessed nerve regeneration and glial scar formation in the injury site using immunofluorescence staining. Neuronal marker Tuj-1 (red fluorescence) and astrocyte marker GFAP (green fluorescence) were used for labeling.

In terms of nerve regeneration, as shown in Fig. 7A, the OC0.33+V3 combined treatment group showed abundant Tuj-1+ positive cells distributed not only in the core injury site but also in the rostral and caudal adjacent tissues. These cells were diverse in shape and densely distributed, suggesting the presence of numerous immature neurons and active neuronal regeneration. In contrast, in the V3 cell transplantation and OC0.33 hydrogel alone groups, only sporadic Tuj-1 positive signals were observed. In the SCI control group, almost no Tuj-1 positive cells were detected, indicating a severe impairment of endogenous nerve regeneration (Fig. 6A). Quantitative analysis of the Tuj-1+ area percentage in the injury site and adjacent regions further showed that the OC0.33+V3 group had a significantly higher positive area ratio compared to the SCI group (Fig. 6B–D).

Fig. 7.

Combinatorial therapy promotes axonal regeneration and remyelination after spinal cord injury. (A) Representative immunofluorescence images of neurofilament 200 (NF200, red) and myelin basic protein (MBP, green) in the lesion area at 16 weeks post-injury. (a-d1) Peri-lesion region (rostral), (a-d2) lesion epicenter, (a-d3) peri-lesion region (caudal). Scale bars: 500 μm (overview), 50 μm (inset). (B–D) Quantification of the NF200-positive area in the (B) rostral, (C) lesion epicenter, and (D) caudal regions (n = 5). (E–G) Quantification of the MBP-positive area in the (E) rostral, (F) lesion epicenter, and (G) caudal regions (n = 5). All data are presented as mean ± SD. Statistical differences were determined using one-way ANOVA. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Fig. 6.

Combinatorial therapy with OC0.33 + V3 promotes axonal regeneration and attenuates glial scar formation after SCI. (A) Representative immunofluorescence images of the axonal marker Tuj-1 (red) and the astrocytic marker GFAP (green) in the lesion site at 16 weeks post-injury. (a-d1) Peri-lesion region (rostral), (a-d2) lesion epicenter, (a-d3) peri-lesion region (caudal). Scale bars: 500 μm (overview), 50 μm (inset). (B–D) Quantification of the Tuj-1-positive area in the (B) rostral, (C) lesion epicenter, and (D) caudal regions (n = 5). (E, F) Quantification of the GFAP-positive area in the (E) rostral and (F) caudal peri-lesion regions (n = 5). All data are presented as mean ± SD. Statistical differences were determined using one-way ANOVA. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Previous studies have indicated that excessive activation and proliferation of astrocytes after SCI are key pathological factors contributing to glial scar formation, which also inhibits functional recovery [52]. The results of this study showed that in the SCI group, widespread and strong GFAP expression was observed at the peripheral injury site, indicating the formation of dense glial scars around the injury. In contrast, the OC0.33 and OC0.33+V3 groups showed fewer GFAP-positive astrocytes at the injury boundary (including rostral and caudal regions), and their arrangement was more loose. This phenomenon may be related to the well-known scar-inhibitory effect of chitosan, the main material of the hydrogel. Further quantitative analysis indicated that the GFAP-positive area at the injury boundary in the OC0.33+V3 group was significantly lower than in the SCI, OC0.33, and V3 groups, suggesting that the combined treatment effectively reduced the accumulation of activated astrocytes (Fig. 6E–F). In summary, the combined application of OC0.33 hydrogel and V3 progenitor cells, through a multi-level synergistic mechanism, effectively promotes neural regeneration while suppressing excessive astrocyte activation. This significantly reduces glial scar formation, creating a favorable microenvironment for neural repair. This dual-track strategy of “material guidance-cell execution” provides crucial theoretical support for developing intelligent spinal cord injury repair solutions capable of simultaneously regulating multiple pathological processes.

2.8. OC0.33+V3 combined treatment promotes axon and myelin regeneration after spinal cord injury

Myelin regeneration and remyelination of regenerating axons are crucial for restoring nerve signal transmission, and they serve as an essential basis for functional recovery after SCI [53]. To evaluate the promotion effect of OC0.33+V3 combined treatment on this process, we performed immunofluorescence staining to mark axon marker neurofilament 200 (NF200) and myelin basic protein (MBP) to evaluate axon regeneration and remyelination.

In the SCI group, only a few NF200+ axon fibers were observed in the injury site and surrounding regions, with sparse morphology and disordered arrangement, indicating severe limitation in axon regeneration, likely related to the inhibitory glial scar microenvironment. In contrast, the OC0.33+V3 combined treatment group showed abundant NF200+ fibers infiltrating both the injury site and surrounding areas, with significantly increased fiber density, continuous morphology, and more organized arrangement (Fig. 7A). Further quantitative analysis of the NF200+ area percentage in the injury site and adjacent regions showed that the OC0.33+V3 group had a significantly higher positive area ratio compared to the SCI group (Fig. 7B–D), suggesting that OC0.33 hydrogel combined with human spinal V3 progenitor cell transplantation can effectively promote axon regeneration.

In terms of myelin regeneration, MBP staining revealed weak and sparse MBP fluorescence signals in the SCI group, indicating severe myelin fragmentation and loss of remyelination capacity. In the OC0.33+V3 group, large numbers of MBP + structures were observed surrounding NF200+ axons in the injury site and surrounding areas, significantly outperforming the other groups (Fig. 7A). The quantitative analysis of the MBP + area percentage in the injury site and adjacent regions was consistent with the above observation, with the OC0.33+V3 group showing significantly higher ratios in the injury center and rostral regions (Fig. 7E–G). In summary, OC0.33 hydrogel combined with V3 progenitor cell transplantation promotes axonal regeneration and myelin repair, enhancing the structural integrity of myelinated fibers for efficient neural signal conduction.

2.9. OC hydrogel scaffold significantly suppresses inflammatory response after spinal cord injury

Neuroinflammation is a central pathological response following SCI, and its degree and duration significantly affect the survival and regeneration of neural tissues [54]. After injury, the blood-spinal cord barrier is disrupted, and in addition to the activation of resident microglia, a large number of peripheral immune cells (such as macrophages and neutrophils) infiltrate the lesion site [55]. These cells release pro-inflammatory cytokines, creating a microenvironment that inhibits regeneration and exacerbates secondary injury. Recent studies have shown that regulating the polarization of microglia/macrophages from a pro-inflammatory M1 phenotype to an anti-inflammatory M2 phenotype is a key strategy to reduce neuroinflammation and promote repair [56].

To evaluate the immune modulation effect of OC0.33 hydrogel and V3 cell transplantation on neuroinflammation after SCI, we performed immunofluorescence staining analysis of the spinal cord tissue at the injury site at 16 weeks post-surgery. The results showed that in the SCI group, numerous IBA1-positive (IBA1+) cells were observed, and a large number co-expressed iNOS, indicating widespread activation of pro-inflammatory M1 cells and a significant inflammatory response. In contrast, in the OC0.33+V3 group, only a few IBA1+ cells were observed, and almost no iNOS + expression was detected (Fig. 8A). Quantitative analysis (Fig. 8C) showed that the relative fluorescence intensity of iNOS in both the OC0.33+V3 and OC0.33 groups was significantly lower than in the SCI group (p < 0.05), indicating that hydrogel implantation effectively suppressed the injury-induced inflammatory response. Notably, we observed a large number of arginase-1 (Arg1)-positive cells in the OC0.33+V3 group, whereas almost no Arg1 expression was detected in the SCI group (Fig. 8B). Quantitative results were consistent with staining results (Fig. 8D). Arg1 is a marker for M2 anti-inflammatory macrophages/microglia, and this result suggests that the combined treatment not only inhibits M1 polarization but also actively promotes the polarization of immune cells toward the M2 phenotype, enhancing their anti-inflammatory and tissue repair functions. In summary, the combined application of OC0.33 hydrogel and human spinal V3 progenitor cells synergistically exerts bidirectional immune modulation: inhibiting M1 polarization while promoting M2 polarization, effectively alleviating neuroinflammation and reshaping the injury microenvironment to create favorable conditions for neuroregeneration.

Fig. 8.

The OC hydrogel significantly suppresses inflammatory response and promotes angiogenesis after spinal cord injury. (A) Representative immunofluorescence images showing IBA1 (microglia/macrophages) and iNOS (an M1 phenotype marker) expression in the lesion area at 16 weeks post-injury across groups. Scale bars: 50 μm. (B) Representative immunofluorescence images of IBA1 and Arg1 (an M2 phenotype marker) expression. Scale bars: 50 μm. (C) Quantitative analysis of the relative fluorescence intensity of iNOS (n = 5). (D) Quantitative analysis of the relative fluorescence intensity of Arg1 (n = 5). (E) Representative immunofluorescence images of CD31 in different groups at 16 weeks post-SCI. White arrows indicate newly formed blood vessels. Scale bar: 100 μm. (F) Quantification of CD31/DAPI area at the injury site of the spinal cord (n = 5). All data are presented as mean ± SD. Statistical differences were determined using one-way ANOVA. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

2.10. OC0.33+V3 combined treatment effectively promotes angiogenesis after spinal cord injury

After spinal cord injury (SCI), the disruption of the local microvascular system leads to a series of pathological responses associated with both primary and secondary damage, severely hindering cell regeneration and repair [57]. Previous studies have demonstrated that the restoration of the microvascular system is crucial for promoting the survival and functional recovery of nerve cells in the damaged area [58]. To evaluate the angiogenic effect of the OC0.33+V3 combined treatment, we assessed blood vessel regeneration in the injury site using CD31 immunofluorescence staining at 16 weeks post-operation.

The immunofluorescence results in Fig. 8E show that, compared to the SCI group, the OC0.33+V3 combined treatment group exhibited a significantly higher density of CD31⁺ blood vessel structures (indicated by white arrows showing newly formed vascular units). This suggests that angiogenesis was more active in the OC0.33+V3 group. Notably, the increased number of neovessels was accompanied by more mature tubular morphology, with branching complexity and lumen diameter distribution more closely resembling the normal spinal cord microvascular network. Quantitative analysis further confirmed this phenomenon: the CD31⁺ area/DAPI area ratio in the OC0.33+V3 group was significantly higher than that in the SCI control group (Fig. 8F). Furthermore, while both OC0.33 hydrogel application and V3 cell transplantation alone demonstrated some improvement in vascular density, their effects were weaker than those observed in the combined treatment group. In summary, these results indicate that OC0.33+V3 combined treatment can effectively promote angiogenesis after spinal cord injury, improve local blood supply, and remodel the microcirculation system of the damaged area, providing a more favorable microenvironment for nerve repair and tissue regeneration.

It is worth noting that even without a hydrogel scaffold, the group receiving V3 cell transplantation alone exhibited certain anti-inflammatory and pro-angiogenic effects. This observation is consistent with reports in the literature: neural progenitor cells inherently possess immunomodulatory and pro-angiogenic properties. Studies have shown that neural progenitor cells can promote the polarization of macrophages toward the anti-inflammatory M2 phenotype by secreting cytokines, thereby reducing inflammation [59]; simultaneously, they can support angiogenesis by secreting factors such as VEGF or through direct cell-cell contact [[60], [61], [62]]. Therefore, V3 cell transplantation alone may also exert tissue repair effects through these endogenous mechanisms, while the addition of a hydrogel may further amplify this effect by increasing cell retention and providing three-dimensional scaffolding.

2.11. OC0.33+V3 combined treatment improves neurogenic bladder structure and function after spinal cord injury

Neurogenic bladder (NGB) is a severe complication following SCI, significantly increasing the risk of urinary tract infections and renal dysfunction, and severely impairing the patient’s quality of life [63]. To systematically assess the impact of OC0.33+V3 combined treatment on SCI-induced bladder dysfunction and pathological tissue changes, we performed macroscopic and histopathological analyses of the bladder at 16 weeks post-operation.

Macroscopic observation revealed that in the SCI group, the bladder displayed typical pathological changes, including significant enlargement, thin walls, and a translucent appearance. In contrast, the bladder in the OC0.33+V3 group was visibly reduced in size (Fig. 9A). Morphometric measurements showed that, compared to the SCI group, the bladder wet weight, bladder/body weight ratio, bladder diameter, and circumference in the OC0.33+V3 group were significantly reduced (all p < 0.05) (Fig. 9B–E), indicating that the combined treatment effectively alleviated the compensatory hypertrophy and urinary retention caused by SCI.

Fig. 9.

Combinatorial therapy ameliorates neurogenic bladder dysfunction and pathological remodeling after SCI. (A) Representative macroscopic images of bladders harvested from each group at 16 weeks post-surgery. Scale bar: 1 cm. (B–E) Quantitative assessments of (B) bladder weight, (C) bladder-to-body weight ratio, (D) diameter, and (E) circumference at the study endpoint (n = 5). (F) Representative H&E-stained sections of the bladder wall. Scale bars: 1 mm (overview); 200 μm (a1-a4); 100 μm (b1-b4). (G) Quantification of bladder wall thickness from H&E-stained sections (n = 5). (H) Representative Masson’s trichrome-stained sections of bladders. Scale bar: 200 μm. (I) Quantitative analysis of the smooth muscle-to-collagen area ratio in the bladder wall (n = 5). All data are presented as mean ± SD. Statistical differences were determined using one-way ANOVA. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Histological analysis further confirmed this improvement at the microscopic level. H&E staining showed that in the SCI group, the detrusor muscle fibers were disorganized with enlarged inter-fiber gaps, and the structure was loose. In contrast, the OC0.33+V3 group exhibited well-organized muscle fibers and a denser structure (Fig. 9F). Quantitative analysis of the bladder wall thickness showed that the OC0.33+V3 and V3 groups had significantly greater bladder wall thickness than the SCI group (Fig. 9G), suggesting that the combined treatment better preserved the structural integrity of the bladder wall. Masson’s trichrome staining to assess fibrosis revealed extensive blue collagen deposition in the bladder’s submucosal layer of the SCI group, with smooth muscle tissue being largely replaced by fibrous tissue. In the OC0.33+V3 group, collagen deposition was significantly reduced, and the smooth muscle structure was better preserved (Fig. 9H). Further quantitative evaluation of the smooth muscle/collagen area ratio showed that the OC0.33+V3 and V3 groups had significantly higher ratios than the SCI group (Fig. 9I), with the combined treatment showing superior effects compared to the individual treatments, confirming that OC0.33+V3 treatment effectively delayed or reversed the pathological fibrosis process in the bladder tissue post-SCI. It should be noted that the primary endpoint of this study was the recovery of hindlimb motor function, and bladder-related assessments were secondary endpoints focusing on structural pathological changes; therefore, urinary parameters such as voiding volume, post-void residual volume, and urodynamic parameters were not measured. The OC0.33+V3 combined treatment strategy promotes spinal cord injury repair by enhancing motor and sensory pathway recovery and protecting against neurogenic bladder dysfunction.

3. Discussion

Although significant progress has been made in recent years in research on spinal cord injury repair, functional recovery remains a major challenge due to the limited regenerative capacity of the central nervous system, the formation of an inhibitory microenvironment following injury, and the insufficient survival and integration of transplanted cells within the lesion site [64,65]. In this study, we developed an injectable, self-healing OHA-CMCS hydrogel and combined it with human spinal cord V3 neural progenitor cells derived from hESCs to establish a composite repair strategy for complete spinal cord transection. V3 interneurons are a class of excitatory interneurons in the spinal cord that participate in the regulation of rhythmic movements (such as walking) and therefore have potential therapeutic value in SCI repair [66]. Our results indicate that this system enhances the retention rate of transplanted cells, supports cell survival, and is associated with improved functional recovery post-transplantation. More importantly, the V3 cell-hydrogel composite suggests that biomaterial-assisted cell therapy may be superior to traditional cell suspension transplantation.

OHA-CMCS hydrogels provide a three-dimensional matrix that offers both structural support and injectability. The hydrogel’s self-healing and shear-thinning properties may reduce mechanical damage during delivery and help maintain structural integrity after implantation. These characteristics are particularly important in spinal cord injury, given the irregular morphology of the injury cavity and the fragility of the surrounding tissues [67,68]. Compared to direct cell injection, encapsulating cells within a hydrogel may enhance local cell retention and reduce early loss of transplanted cells caused by leakage, mechanical stress, and rapid exposure to the inflammatory microenvironment [69]. In this sense, the hydrogel serves not merely as a passive carrier but as a biomaterial scaffold capable of improving the feasibility of neural progenitor cell transplantation in highly adverse tissue environments. In addition to the aforementioned physical support function, hydrogels themselves may possess capabilities that extend beyond those of a passive carrier. Their components—hyaluronic acid and chitosan—possess good biocompatibility and potential immunomodulatory activity, which may exert a positive influence on the repair environment [70]. However, it remains unclear whether their effects stem from the biochemical activity of the materials themselves, enhanced cell retention, the local retention of cell-secreted factors, or a combination of multiple mechanisms.

Furthermore, the V3 cell-hydrogel system is closely associated with favorable changes in the injured microenvironment. Specifically, animals implanted with the V3-hydrogel system exhibited characteristics such as reduced glial scarring, attenuated inflammatory responses, and enhanced vascular-related signaling in their injured microenvironment. Although these results are consistent with the view that the combined therapy modulates the tissue environment of the injury site, this study did not directly elucidate the individual contributions of each component nor identify the specific molecular pathways involved. Therefore, a more appropriate conclusion is that the combined therapy is associated with a microenvironment more conducive to repair, rather than asserting that it achieves explicit microenvironmental reprogramming. Future studies should employ methods such as conditioned medium assays, cytokine profiling, spatial transcriptomics, and targeted knockout approaches to further elucidate how the hydrogel matrix and V3 neural progenitor cells synergistically influence astrocytes, microglia, endothelial cells, and other microenvironmental components following transplantation [71].

The improvements in the aforementioned microenvironment created conditions conducive to the survival and integration of transplanted cells. Building on this, we further validated the differentiation potential of V3 neural progenitor cells within the hydrogel. This study found that V3 progenitor cells derived from hESCs can be obtained through a staged differentiation protocol and retain their developmental potential even after being encapsulated in the hydrogel. The cells generated by this differentiation strategy express markers consistent with the V3 lineage, supporting the feasibility of preparing spinal cord intermediate neuron precursors for transplantation. In vivo, we detected the transplanted human cells at the injury site, indicating that the hydrogel can support short-to medium-term survival of the graft in the harsh injury environment. However, the current data do not yet fully clarify the maturation status, homogeneity, and long-term fate of these cells following implantation. In particular, further transcriptomic analysis and longer-term phenotypic analysis will help to more reliably assess their lineage characteristics and differentiation fidelity. Similarly, future studies should more comprehensively evaluate whether the transplanted cells acquire stable, mature V3-like characteristics and how their in vivo status evolves over time.

At the functional level, we observed improvements in both motor behavior and electrophysiological parameters in nude rats following treatment with the V3 cell-hydrogel composite system. These encouraging results suggest that the transplanted cells and the biomaterial matrix jointly promote functional recovery. In this study, we further employed chemogenetics to selectively inhibit the activity of transplanted V3 neurons; the results showed a significant decline in motor function in rats following inhibition, providing direct evidence that transplanted neurons contribute to the process of functional recovery. However, chemogenetics experiments alone cannot fully elucidate the synaptic architecture or the synaptic pathways involved between the host and the graft. Therefore, although our data support graft-mediated functional integration, direct verification of synaptic connections and circuit-level repair is still required through experiments such as transsynaptic tracing, optogenetic circuit analysis, and co-staining of synaptic markers [72].

The optimized transplantation parameters were critical to the functional recovery observed in this study. Regarding cell dose, the dose of 1 × 10⁶ cells per animal used in this study was determined based on preliminary dose-response results (comparing 4 × 10⁵, 8 × 10⁵, and 1 × 10⁶ cells) and is consistent with the optimal dose reported in previous studies [73]. Regarding the time window, this study opted for acute-phase transplantation (injection immediately following injury), which allows the hydrogel to maximize its hemostatic, neuroprotective, and physical scaffolding effects; however, the clinical feasibility of the acute window is limited, and future research should explore subacute or chronic-phase transplantation strategies. Regarding the administration method, the injectable system with in situ gelation enables localized, targeted delivery, improves cell survival and retention rates, and involves minimally invasive procedures, demonstrating potential for clinical translation.

In addition to cell replacement, paracrine effects may also be a key mechanism underlying recovery. It has been reported that neural progenitor cells release various neurotrophic and immunomodulatory factors, thereby promoting neuroprotection, angiogenesis, and tissue repair [74,75]. In this study, the group receiving V3 transplantation alone also exhibited some degree of functional improvement, suggesting that V3 may have independent beneficial effects. However, since we have not yet directly analyzed the secretome or the effects of conditioned medium, this mechanism requires further validation. Future studies could utilize proteomics, cytokine profiling, and conditioned medium systems to assess whether V3 progenitor cells regulate inflammation, vascular remodeling, or the glial response following SCI via paracrine pathways.

This study also has several limitations. First, all animal experiments were conducted in immunodeficient animals, so the impact of immune rejection on transplantation outcomes—a critical factor for clinical translation—has not yet been assessed. Second, while the complete transection injury model is convenient for mechanistic studies, it differs from the contusion-type spinal cord injury more commonly encountered in clinical practice. Third, the long-term degradation behavior and metabolic safety of the hydrogel have not yet been fully elucidated. Fourth, this study primarily focused on motor function and selected histological endpoints; however, the assessment of sensory and autonomic function, as well as long-term safety, remains insufficient. Fifth, the cells used in this study were derived from hESCs; although no tumor formation was observed during the short-term follow-up, the potential risks of tumorigenicity and immunogenicity still require further evaluation in studies with larger sample sizes and longer observation periods.

In summary, this study demonstrates that combining injectable, self-healing hydrogels with human spinal cord V3 neural progenitor cells derived from hESCs may offer a promising strategy for SCI repair. This composite system may exert its effects by increasing the retention rate of transplanted cells, improving the local repair environment, and promoting functional recovery. Future work should involve more systematic mechanistic studies and long-term safety evaluations to further optimize material composition, cell dosage, and transplantation strategies, thereby advancing its translation into clinical practice.

4. Methods

4.1. Synthesis of oxidized hyaluronic acid (OHA)

Hyaluronic acid (HA) with a molecular weight of 6 × 10⁵ was oxidized using sodium periodate (NaIO₄) to introduce aldehyde groups (-CHO) at the -OH positions. Specifically, 2.0 g of HA was dissolved in 200 mL of PBS buffer, and then 1.0 g of sodium periodate was added. The oxidation reaction was allowed to proceed in the dark for 8 h, followed by the addition of 2 mL of ethylene glycol, which was stirred for an additional 30 min to terminate the reaction. The final product was purified by dialysis for 3 days and then freeze-dried to obtain oxidized hyaluronic acid (OHA).

4.2. Preparation of OHA-CMCS hydrogel

The freeze-dried OHA and CMCS were dissolved in deionized water to prepare 3% (w/v) OHA and 3% (w/v) CMCS solutions. Hydrogels were then synthesized by mixing these solutions in different volumetric ratios (OHA: CMCS = 3:1, 2:1, 1:1, 1:2, 1:3), referred to as OHA-CMCS hydrogels and abbreviated as OC hydrogels. For convenience, the hydrogels with different volumetric ratios were named OC3, OC2, OC1, OC0.5, and OC0.33, where the suffix represents the volumetric ratio of OHA to CMCS. Gelation of the hydrogels was based on the Schiff base reaction between the -CHO and -NH2 groups.

4.3. Characterization of hydrogel properties

The internal and cross-sectional three-dimensional structure of the hydrogels was examined by scanning electron microscopy (SEM, Hitachi SU8010, Japan). In brief, hydrogel samples were dried in liquid nitrogen and then shattered. Prior to the experiment, the samples were coated with platinum for 60 s under vacuum. The working conditions were low acceleration voltage (5 kV) and short working distance (6.0 mm). Fourier transform infrared (FTIR) spectroscopy was used to analyze the FTIR spectra of OHA, HA, CS, and CMC. The samples were freeze-dried before testing. Background spectra were collected before obtaining the transmission infrared spectra of the samples. All spectra were obtained between 4000 cm⁻¹ and 500 cm⁻¹. The rheological properties of the hydrogels, including storage modulus (G′) and loss modulus (G″), were studied using a rheometer (AntonPar, MCR-102, Austria).

4.4. In vitro cytotoxicity of the hydrogel

Based on previous studies [76], the hydrogel samples were modified slightly. Briefly, sterilized hydrogel samples were soaked in SH-SY5Y cell complete culture medium (Haixing Biotechnology, TCH-G325) at a concentration of 0.1 g/mL. The hydrogel was extracted at 37 °C for 24 h and then removed. The culture medium was filtered through a 0.22 μm filter.

SH-SY5Y cells were seeded at 3000 cells/well in a 96-well tissue culture plate and treated with the hydrogel extract and untreated complete medium as the control. Each group was set up with five replicate wells, and at day 1, 2, and 3, the medium was replaced with PBS buffer, washed three times, and then 100 μL of medium and 10 μL of CCK-8 working solution were added. After incubating for 1 h at 37 °C in the dark, absorbance at 450 nm was measured using a microplate reader. Cell viability was calculated using the following formula:

$$cellviability(\%)=\frac{{OD}_{sample}-{OD}_{cck-8}}{{OD}_{control}-{OD}_{cck-8}}\times 100\%$$

For further validation of cytocompatibility, live/dead staining was performed. SH-SY5Y cells were seeded at 10^4 cells/well on confocal culture dishes (Biosharp, BS-20-GJM). After 12 h, the cell culture medium was replaced with the hydrogel extract of different samples. After 24 h of incubation, Calcein-AM/PI live/dead staining kit was used to detect the SH-SY5Y cell morphology. Laser confocal microscopy (Olympus, FV300) was used for observation and recording.

4.5. Effects of hydrogel on the viability of human spinal V3 progenitor cells

Human spinal cord V3 progenitor cells were harvested on day 21 of a four-stage differentiation protocol. At this stage, the cells were cultured in DMEM/F12 neural expansion medium supplemented with B27, EGF, and FGF-2, where they spontaneously formed glutamatergic neurospheres. Prior to encapsulation, the neurospheres were harvested by gentle centrifugation and treated with Accutase for 2–3 min to partially disaggregate them into small cell clusters rather than completely dissociated single cells. The partially disaggregated neurospheres were mixed with a hydrogel precursor solution and then seeded onto Matrigel-coated cover slips. The control group was cultured in complete medium. The growth of human spinal cord V3 progenitor cells was observed at 1, 3, 5, and 7 days of co-culture using an inverted optical microscope (Olympus, CKX41).

After 7 days of co-culturing, the cells on the coverslips were fixed with 4% paraformaldehyde (PFA) (Biosharp, BL539A) at room temperature for 30 min. The samples were then washed three times with PBS, each wash lasting 5 min, followed by blocking with Quick BlockTM blocking buffer (Beyotime, P0260) for 15 min. Primary antibodies were added in Quick BlockTM Primary Antibody Dilution Buffer (Beyotime, P0262) and incubated overnight at 4 °C. The following day, the sections were washed three times with TBST, each wash lasting 5 min, followed by incubation with fluorescently labeled secondary antibodies and Hoechst 33342 (Invitrogen, H1399) for 1 h at room temperature in the dark. Finally, the coverslips were washed three times with PBS, each wash lasting 5 min, and then mounted on slides after drying.

The primary antibodies used were: Rb VGLUT2 (1:500, Synaptic Systems, 135403) and Ms MAP2 (1:500, Sigma, M1406). The secondary antibodies used were: Hoechst 33342 (1:1000, Invitrogen, H1399), Donkey anti-rabbit Alexa Fluor 488 (1:1000, JACKSON, 711-545-152), and Donkey anti-mouse Alexa Fluor 594 (1:1000, JACKSON, 711-585-150). All samples were visualized using a laser scanning confocal microscope.

4.6. Spinal cord transection and hydrogel implantation surgery

Six to eight-week-old female nude rats (weighing 110-160 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. After arrival, the rats were acclimatized in the animal facility for at least one month before the experiments began. Each animal was housed individually in an independent ventilated cage (IVC; height 30 cm, width 27 cm, depth 40 cm) in a specific pathogen-free (SPF) room with a temperature of 22-24 °C, humidity of 45-60%, and a 12-h light/dark cycle (lights on at 6:00 a.m. and off at 6:00 p.m.). During the experiment, animals had ad libitum access to food and water.

Prior to surgery, the body weight of each nude rat was measured and recorded. Anesthesia was induced via intramuscular injection of Zoletil 50 (Virbac S.A., Carros, France; 10 mg/kg) and Xylazine (2.5 mg/kg). To prevent infection, ceftriaxone (50 mg/kg) was administered intraperitoneally, and erythromycin ointment was applied to prevent dry eye. The rats were placed in a prone position on a heating pad maintained at 37 °C. The hair around the T10 spinous process was shaved, and the surgical area was thoroughly disinfected with povidone-iodine. A #11 scalpel was used to make an incision along the midline of the back, followed by the separation of surrounding muscles and fascia using surgical scissors. The lamina was removed to expose the dorsal aspect of the T9-T10 segments. The spinal cord was transected using Vannas scissors to create a 2.0 mm gap. After complete hemostasis, OC0.33 hydrogel, human spinal V3 progenitor cells, and OC0.33 + human spinal V3 progenitor cells were implanted into the lesion site. The SCI group, which received no treatment, served as the control. Finally, the muscles and skin were sutured. Post-surgery, the rats were weighed daily, and the wound was disinfected with povidone-iodine. To prevent dehydration and infection, 5 mL of physiological saline was injected subcutaneously, and ceftriaxone was administered intraperitoneally. Bladder massage was performed twice daily to assist urination for nearly two months.

4.7. Cell and hydrogel preparation and transplantation

V3 neural progenitor cells at day 21 of differentiation were used for in vivo transplantation experiments. At this stage, the cells had formed neural spheroids. To obtain cell clusters suitable for transplantation, the day-21 neural spheroids were gently treated with Accutase at 37 °C for 3–5 min, followed by gentle pipetting to partially dissociate the large neural spheroids into smaller ones, rather than completely dissociating them into single cells. In this study, 1 × 10⁶ human spinal cord V3 neural progenitor cells were used per animal; this dose was determined based on preliminary experiments and is consistent with the optimal dose identified in previous rat models of spinal cord injury following comparisons of 4 × 10⁵, 8 × 10⁵, and 1 × 10⁶ cells [73].

OC0.33 hydrogel, human spinal cord V3 precursor cells, and the OC0.33 + human spinal cord V3 precursor cell composite were injected immediately (within 0 h) following spinal cord transection. For complex transplantation, human pluripotent stem cell-derived V3 neural progenitor cells were suspended in the CMCS precursor solution of the OHA-CMCS hydrogel. Immediately prior to injection, the suspension was mixed with the OHA solution and delivered directly to the site of spinal cord injury via a syringe. Following injection, the hydrogel gelled in situ at the injury site, enabling localized, targeted delivery.

4.8. Behavioral assessment

Motor function was evaluated before and after spinal cord injury (SCI). The motor function of rats was assessed from multiple dimensions: hind limb function was evaluated using the Basso, Beattie, and Bresnahan (BBB) open-field locomotor rating scale. The horizontal ladder test was used to assess inter-limb coordination, and the hind limb grip strength test evaluated hind limb strength. Additionally, the rats’ walking patterns were evaluated through video analysis of their hind limb movements. All results were measured by two independent observers who were blinded to the group identity.

4.9. Electrophysiological examination

Briefly, the animals were anesthetized with isoflurane gas (2.5-3%), and the right hind limb was shaved and cleaned. Bipolar needle electrodes were used to stimulate the sciatic nerve, with proximal stimulation near the sciatic notch and distal stimulation close to the knee. The recording electrode was placed on the gastrocnemius muscle to record compound muscle action potentials (CMAPs), with the ground electrode placed on the tail. The stimulation intensity was set to 3 mA, with three repetitions for each rat.

4.10. Tissue processing

Before tissue processing, nude rats were anesthetized, and their hearts were perfused with heparinized (100,000 U/L) physiological saline, followed by 4% paraformaldehyde (PFA) perfusion. The brain and spinal cord were collected and fixed in 4% PFA at 4 °C overnight, followed by immersion in 30% sucrose solution for 72 h. Spinal cord segments containing cells (T8-T12) were dissected, embedded in optimal cutting temperature (OCT) compound (Sakura, 4853), and rapidly frozen. Using a cryostat (Thermo Fisher, CRYOSTAR NX50), 20 μm thick horizontal sections were cut and stored in a cryoprotectant solution (PBS solution containing 30% sucrose and 30% ethylene glycol) at −80 °C.

4.11. Immunofluorescence

Spinal cord tissue samples were fixed at room temperature for 30 min using 4% paraformaldehyde (Servicebio, China), followed by thorough washing with phosphate-buffered saline (PBS) to remove residual fixative. To block nonspecific binding and enhance antibody penetration, the samples were incubated at room temperature for 1 h in a blocking solution containing 0.3% Triton X-100 (Sigma, Germany) and 5% bovine serum albumin. After blocking, the samples were incubated overnight at 4 °C with specific primary antibodies at the following dilutions: Tuj-1 (1:200), GFAP (1:500), MBP (1:200), iNOS (1:100), Arg-1 (1:200), NF200 (1:200), STEM121 (1:500), and NeuN (1:500). The primary antibodies were sourced from Servicebio (China), except for STEM121 (Takara, Y40410) and NeuN (Millipore, ABN78). The next day, the samples were washed with PBS, then incubated for 1 h at room temperature in the dark with the corresponding fluorophore-conjugated secondary antibodies: CY3-labeled donkey anti-mouse IgG (1:1000) or Alexa Fluor 488-labeled donkey anti-rabbit IgG (1:1000), respectively (Servicebio, China). Finally, the samples were counterstained with DAPI (1:1000; Servicebio, China) for 10 min. All fluorescence images were captured using the 3DHISTECH digital slide scanning system and analyzed semi-quantitatively using ImageJ software.

4.12. Statistical analysis

All graphs were generated using Origin and GraphPad Prism (version 10.0). Experimental data are presented as the mean ± standard deviation. Statistical differences between groups were evaluated using GraphPad Prism software. An independent t-test was used for comparisons between two groups, while one-way ANOVA was used for comparisons among multiple groups, followed by post hoc Bonferroni correction. Statistical significance was set at ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001; ns indicates no statistical difference.

Ethics approval and consent to participate

All animal experiments conducted in this study were approved by the Institutional Animal Care and Use Committee (IACUC) of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (Approval No. T1-2024-06-070) and strictly complied with the Guidelines for Laboratory Animal Welfare and Ethics as well as relevant laws and regulations.

CRediT authorship contribution statement

Wenqi Yin: Data curation, Formal analysis, Investigation, Validation, Writing – original draft, Writing – review & editing. Guangrui Ma: Formal analysis, Writing – review & editing. Jia Xu: Formal analysis, Investigation. Xinrong Chen: Investigation, Validation. Yupeng Liu: Investigation. Qiuzhi Zhou: Formal analysis, Writing – review & editing. Guoliang Tang: Writing – review & editing. Zhijun Shi: Conceptualization, Funding acquisition. Guang Yang: Funding acquisition, Supervision. Hong Chen: Conceptualization, Funding acquisition, Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: