Dual-engineered Treg-Exosome-IKVAV nanovesicles spatiotemporally sequentially regulate neuro-immune microenvironment to promote spinal cord injury repair

Abstract

Spinal cord injury (SCI) repair remains a significant clinical challenge due to the imbalance of inflammatory microenvironment and insufficient neural regenerative ability. Current therapeutic approaches, such as pharmaceuticals, stem cell transplantation, and inorganic biomaterials, are limited by insufficient supply, poor bioactivity, and immunogenicity, severely limiting their clinical translation. To address these issues, we developed a biomimetic dual-engineered nanovesicles (Treg-Exo-IKVAV) by conjugating a neuroprotective IKVAV peptide motif with Treg-derived exosomes (Treg-Exo) using click chemistry. This system synergistically integrates the intrinsic immunomodulatory properties of Treg-Exo (early-stage anti-inflammation) with the neural regenerative capability of IKVAV, enabling spatiotemporally sequential regulation of neuro-regeneration. In vitro studies demonstrated that Treg-Exo-IKVAV suppressed macrophage-induced inflammatory responses by reprogramming macrophage polarization. Furthermore, Treg-Exo-IKVAV exerts dual direct and immunoregulatory effects on promoting neuronal differentiation of stem cells (NSCs). In vivo experiments revealed that Treg-Exo-IKVAV via tail vein injection precisely targeted and accumulated at the injured site. Subsequently, the functional assessments showed that Treg-Exo-IKVAV significantly enhanced motor functional recovery in SCI mice. Mechanistically, these nanovesicles reshaped the neuro-immune microenvironment through a two-phase mechanism: initial suppression of inflammation via Treg-derived anti-inflammatory signaling followed by activation of neuro-regenerative pathways mediated by IKVAV. This integrated “exosome-peptide” nanocomposite combining immunomodulation and neuronal regeneration provides a highly efficient and safe therapeutic solution for SCI.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03999-3.

Highlights

SCI repair remains a challenge due to the neuro-immune dysfunction and limited neuro-regeneration capability.

Treg-Exo-IKVAV nanovesicles showed immunomodulatory property, regulating macrophage phenotype transition toward M2 phenotype.

Treg-Exo-IKVAV nanovesicles showed potent dual effects in both direct NSCs differentiation and immune-mediated NSCs differentiation in vitro.

Treg-Exo-IKVAV nanovesicles spatiotemporally and sequentially regulated neuro-immune microenvironment to promote SCI repair in vivo.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03999-3.

Introduction

Spinal cord injury (SCI) is one of the leading causes of mortality and long-term disability worldwide, which triggers a cascade of detrimental events initiated by primary mechanical trauma, followed by secondary inflammatory responses and excitotoxicity [1, 2]. The acute phase is characterized by robust infiltration of neutrophils and macrophages, releasing pro-inflammatory cytokines such as TNF-α and IL-6, which exacerbate tissue damage through oxidative stress and blood-brain barrier disruption [3]. Meanwhile, the chronic phase witnessed persistent microglial activation and astrocytic gliosis, forming inhibitory glial scar that physically blocks axonal regeneration [4]. Neurological deficits after SCI arise from two interconnected mechanisms: (1) Immune dysregulation: Uncontrolled inflammation destroys residual neural tissue and creates a hostile immune microenvironment impeding repair, (2) Insufficient nerve regeneration capacity: Reduced expression of growth factors like brain-derived neurotrophic factor (BDNF) and glial cell line-derived neurotrophic factor (GDNF), coupled with elevated levels of inhibitory molecules, suppresses neural stem cells (NSCs) differentiation and axonal sprouting [5, 6]. Traditional treatments such as corticosteroids and methylprednisolone show limited efficacy in resolving these dual pathologies [7, 8]. Furthermore, the emerging stem cell therapies face challenges including immune rejection and poor engraftment efficiency [9, 10]. Therefore, developing a safer and more effective therapeutic approach that can orchestrate immune homeostasis while stimulating endogenous neurogenesis are urgently needed for SCI repair.

Recent advances in neuroimmunology have elucidated the pivotal interplay between immune cells and neuronal regeneration [11]. Emerging evidence highlights that precise modulation of inflammatory responses profoundly influences neuro-regenerative trajectories, as unresolved inflammation impairs neuronal stem cells (NSCs) differentiation and disrupts repair processes [12]. Among immune regulators, regulatory T cells (Tregs) have garnered attention for their dual roles in maintaining immune homeostasis and promoting tissue repair [13]. Tregs, a subset of T cells with significant immunosuppressive effects, inhibit the proliferation and function of effector T cells through the release of immunosuppressive cytokines and cell-contact dependent mechanisms, thereby maintaining immune homeostasis in the body and preventing excessive immune responses and autoimmune diseases [14–16]. Several studies have reported that Tregs could inhibit the macrophage-induced inflammation by modulating macrophage polarization from pro-inflammatory M1 to anti-inflammatory M2 phenotypes, thereby promoting neural recovery after SCI [13]. Furthermore, research has also found that Tregs were significantly involved in oligodendrocyte differentiation and myelination, thus promoted the functional recovery of SCI [17]. These evidences highlighted the critical role of Tregs in the pathophysiological process after SCI. More importantly, the natural immunosuppressive effects make them ideal candidates for SCI immunotherapy. However, direct cell transplantation therapy still faces various risks, such as poor survival rates and immune rejection, and an easier and safer therapy is urgently needed.

Recently, exosomes, nanoscale extracellular vesicles secreted by various cell types, have emerged as alternatives to direct cell transplantation in several diseases including SCI repair [18]. They exhibit similar functions as parental cells due to the inside bioactive molecules derived from the original cells for cell-to-cell communication, such as mRNAs, miRNAs, and proteins [19, 20]. Furthermore, compared with cell transplantation, exosomes also showed several advantages, such as better stability, lower immunogenicity, and a natural targeting capability [21, 22]. Exosomes derived from macrophages, Schwann cell, BMSCs or NSCs have been demonstrated to effectively promote SCI repair [23]. However, in-depth studies have revealed that despite their neural regenerative potential, these exosomes face limitations in clinical applications due to insufficient specificity in immune regulation. Against this backdrop, Treg-derived exosomes (Treg-Exo), enriched with immunomodulatory cargo such as IL-10, TGF-β, and miR-146a, exhibit unique advantages [14]. Notably, recent studies had reported the Treg-Exo not only effectively inhibited inflammatory cascades but also reprogramed macrophages into a pro-regenerative M2 phenotype, creating an immune microenvironment conducive to SCI repair [24, 25]. Compared to other exosomes, the dual functionality of immunomodulation and regeneration enhancement positions Treg-Exo as a promising candidate for broader applications in the field of neuronal regeneration. Concurrently, the IKVAV peptide (Ile-Lys-Val-Ala-Val), a functional motif within laminin, had been demonstrated significant regulatory capacity on NSCs viability, proliferation, and differentiation [26]. A study found that IKVAV-functionalized hydrogels effectively suppressed glial scar formation while promoting tissue regeneration following spinal cord transection in rats [27]. This evidence highlights the neuro-regenerative ability of IKVAV peptide motifs.

In this study, we proposed a bioinspired spatiotemporally sequential therapy: a dual-functional nanotherapeutic system (Treg-Exo-IKVAV) constructed by covalently anchoring IKVAV peptides onto surface of Treg-Exo via click chemistry (Scheme 1). This design leverages the CD47-mediated “don’t eat me” signaling of Treg-Exo to evade immune clearance, while its immunomodulatory cargo suppresses early-phase inflammatory storms. Subsequently, IKVAV activates the JAK-STAT signaling pathway in NSCs within the acidic injury microenvironment, thereby promoting neuronal differentiation for NSCs. Through comprehensive in vitro and in vivo evaluations in a mouse SCI model, we demonstrate the system’s efficacy in neural repair and elucidate its immunometabolic reprogramming mechanisms. By integrating contemporary neuroimmunology principles with advanced nanotechnology, this dual-engineered “Exosome-Peptide” platform overcomes critical limitations of existing therapeutic strategies, offering a transformative paradigm for clinical neural regeneration.

Fig. 1.

Schematic illustration of the fabrication and evaluation of Treg-Exo-IKVAV for traumatic SCI.(a) The isolation of Treg-Exo and the preparation process of Treg-Exo-IKVAV. (b) Mechanism underlying Treg-Exo-IKVAV’s immunoregulatory and neuro-regenerative capabilities. (c) Treg-Exo-IKVAV spatiotemporally sequentially reshaped the neuro-immune microenvironment, and then promoted functional recovery of SCI mice

Results and discussion

The characterization of Treg-Exo-IKVAV

Recently, increasing studies have demonstrated that Tregs and its exosomes play critical regulatory roles in SCI repair [14]. However, their limited intrinsic neuroregenerative capacity substantially constrains therapeutic efficacy in promoting neural regeneration. Inspired by the excellent neuro-regenerative capability of IKVAV peptide motifs, we engineered the Treg-Exo with IKVAV peptide motifs to endow it with the dual immunomodulatory and neural-regenerative functionalities.

Figure 2a illustrated the fabrication process of Treg-Exo-IKVAV. A two-step conjugation protocol was used for the functionalization of Treg-Exo with IKVAV [28]. First, DBCO-sulfo-NHS was incubated with Treg-Exo to covalently anchor DBCO groups onto exosome surfaces via NHS-mediated amine coupling with phosphatidylethanolamine or membrane proteins. Then, azido-IKVAV peptides were immobilized on the surface of Treg-Exo through DBCO-azide click chemistry. The transmission electron microscope (TEM) was used to observe the morphologies of Treg-Exo and Treg-Exo-IKVAV. As shown in Fig. 2b, both Treg-Exo and Treg-Exo-IKVAV exhibited a cup-shaped morphology. Subsequently, the particle size distributions of Treg-Exo and Treg-Exo-IKVAV were detected by NTA (Fig. 2c). The highest peak of Treg-Exo was approximately 70 nm, while the profile of Treg-Exo-IKVAV showed a peak around 120 nm, indicating that surface modification with IKVAV peptide motifs increased the diameter of Treg-Exo. The polydispersity index (PDI) was further determined to assess the size distribution profile of the exosomes. Treg-Exo and Treg-Exo-IKVAV exhibited PDI values of 0.30 ± 0.03 and 0.38 ± 0.04, respectively. Additionally, the expression of exosome markers, including CD9, TSG101, CD63 and CD81 in the cells, Treg-Exo and Treg-Exo-IKVAV groups was detected by western blot (Fig. 2d and Figure S1). The results verified the successful isolation of Treg-Exo. The flow cytometry quantification of CD9, CD63 and CD81 (Fig. 2e) was performed, and the results further confirmed the successful isolation of Treg-Exo. The quantitative analysis presented in Fig. 2f showed that there was significant difference between Treg-Exo and Treg-Exo-IKVAV in the average diameters, consistent with abovementioned results. However, the zeta potential (Fig. 2g) showed no significant difference between the Treg-Exo and Treg-Exo-IKVAV groups. Taken together, these results indicated that click chemistry-mediated IKVAV motif conjugation specifically targets functional domains without globally altering the Treg-Exo membrane integrity.

Fig. 2.

Preparation and characterization of Treg-Exo-IKVAV. (a) Schematic illustration of the fabrication of Treg-Exo-IKVAV. (b) Typical TEM images of Treg-Exo and Treg-Exo-IKVAV. (c) Particle size distributions of Treg-Exo and Treg-Exo-IKVAV measured by NTA. (d) Exosome markers (CD9, TSG101, CD63, and CD81) detected Western Blot. (e) Exosome markers (CD9, CD63, and CD81) detected flow cytometry. (f) The average diameters of Treg-Exo and Treg-Exo-IKVAV (n = 5). (g) Zeta potentials of Treg-Exo and Treg-Exo-IKVAV (n = 5). (h) The representative fluorescence images of Treg-Exo and Treg-Exo-IKVAV. Treg-Exo was stained by DiO and IKVAV was labeled by Cy5.5. (i-j) The physicochemical stability of Treg-Exo and Treg-Exo-IKVAV after a 7-day incubation period in phosphate-buffered saline (PBS) using time-dependent measurements of particle size distribution and zeta potential values (n = 3). Data are expressed as mean ± SD. Statistical significance was tested by one-way ANOVA (f, g) with Tukey’s multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001

Currently, more and more surface modification techniques have been used for exosomes, such as genetic engineering, physical manipulation, and chemical conjugation, because cell engineering of parent cells is complex and expensive, and transfection rate is low [28, 29]. The physical processing methods like extrusion, freeze-thaw cycles, and sonication risk degradation and leakage of bioactive components inside exosomes. Therefore, there exists a critical need for a safe, efficient, and biocompatible strategies for IKVAV conjugation onto exosomal surfaces. Click chemistry emerges as a promising solution, which utilized bioorthogonal reactions that form stable covalent bonds under physiological conditions through biocompatible reagents [30]. This modular system has demonstrated feasibility in exosome labeling applications [31]. Notably, its inherent chemo-selectivity enables precise modification of biomolecules including nucleic acids and polypeptides without compromising structural integrity [32]. Taken together, click chemistry is an efficient, low-cost, and safe method to conjugate IKVAV onto surfaces of Treg-Exo, and thereby was used in this study.

To confirm the successful conjugation of IKVAV peptide motifs, Treg-Exo was stained by DiO, and the IKVAV was labeled by Cy5.5. Next, the fluorescence of Treg-Exo and Treg-Exo-IKVAV was captured by confocal laser scanning microscopy (CLSM). As shown in Fig. 2h, the co-localization of green (Treg-Exo) and red (IKVAV peptides) fluorescence signals was observed on Treg-Exo-IKVAV, confirming successful functionalization. Co-localization between Treg-Exo and IKVAV was further quantified using ImageJ, yielding a Pearson’s r value of 0.894, indicative of strong spatial association. Additionally, as we mentioned above, the traditional physical engineering methods like extrusion, freeze-thaw cycles, and sonication risk degradation and leakage of bioactive components inside exosomes. Therefore, the stability of Treg-Exo-IKVAV was also evaluated. As shown in Fig. 2i and j, the particle size and zeta potential of Treg-Exo and Treg-Exo-IKVAV remain relative stable over a 7-day period after incubation in PBS, indicating that click chemistry is a safe engineering method for exosomal surface modification.

In vitro Immunomodulatory effects of Treg-Exo-IKVAV

A balanced inflammatory response forms the biological basis for tissue repair, in which the immune system plays a critical regulatory role [33]. However, hyperactivated or dysregulated immune responses may initiate pathological inflammatory cascades, leading to microenvironmental disturbances that impede nerve tissue regeneration. This mechanism underscores the critical importance of orchestrating synchronized immunomodulation in developing functional SCI repair materials. Within the immune cellular network, macrophages act as central effectors, which can be dynamically polarized into pro-inflammatory (M1) or anti-inflammatory (M2) phenotypes subject to different environmental stimuli. M1 subtype exacerbates inflammation and tissue damage, whereas M2 subtype facilitates neuroinflammation resolution and tissue regeneration. After SCI, the macrophages were recruited and polarized into M1 phenotype impeding neural regeneration. Especially, unlike injuries to soft tissues such as the skin and peripheral nerves, SCI lack effective inflammation regulation. This leads to a predominance of the M1 phenotype in macrophages, creating a persistent inflammatory environment that further hinders neural tissue repair and functional recovery. More importantly, the negative regulatory mechanisms of the immune system are suppressed, further exacerbating neuroinflammation. Promoting macrophage phenotype transition from M1 to M2 is emerging as a promising therapeutic strategy to improve neurological outcomes after SCI [34, 35]. Enlightened by this, the immunoregulatory effects of Treg-Exo-IKVAV nanocomposite were systematically investigated, with particular focus on macrophage polarization dynamics, to elucidate its potential role in coupling immunoregulatory and neuro-regenerative processes during SCI repair.

In this part, macrophage cell line Raw 264.7 was used and the activation was induced by lipolyaccharide (LPS). First, to assess RAW 264.7 cell uptake of Treg-Exo and Treg-Exo-IKVAV, exosomes were labeled with FITC (green) and nuclei with DAPI (blue). As demonstrated in Fig. 3a, green-fluorescent signals localized to macrophage cytoplasm, confirming successful internalization of both Treg-Exo and Treg-Exo-IKVAV by target cells. These findings verified the targeting efficacy of the nanovesicle system at the subcellular level.

Fig. 3.

In vitro immunomodulatory effects of Treg-Exo-IKVAV.(a) Uptake of green, fluorescent FITC-labeled Treg-Exo and Treg-Exo-IKVAV into Raw 264.7 macrophages. (b) The typical images of iNOS (green) and CD206 (red) immunostaining. (c) Quantitative analysis of immunostaining (n = 5). (d) Flow cytometry analysis of CD86 and CD206 positive cells after Treg-Exo and Treg-Exo-IKVAV treatments. (e) Quantification of the flow cytometry results (n = 5). (f) RT-PCR detection of macrophage polarization and inflammation-related gene expression (n = 5). (g) KEGG enrichment analysis. (h) GO analysis. (i) Schematic illustration of the immunoregulatory mechanism of Treg-Exo-IKVAV on macrophage. Data are expressed as mean ± SD. Statistical significance was tested by one-way ANOVA (f) or two-way ANOVA (c, e) with Tukey’s multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001

Subsequently, the iNOS/CD206 immunofluorescence staining was performed to evaluate the immunomodulatory effects of Treg-Exo-IKVAV on macrophage polarization dynamics. Figure 3b and c demonstrated that LPS treatment significantly upregulated iNOS expression (M1 macrophage marker) compared with the control group. Treatment with Treg-Exo and Treg-Exo-IKVAV substantially reduced the proportion of iNOS-positive cell while elevating CD206-positive cell abundance compared to the LPS group. However, no phenotypic distinction emerged between Treg-Exo and Treg-Exo-IKVAV groups, indicating that surface modification with IKVAV peptides did not change the immunomodulatory effects of Treg-Exo, highlighting its advantageous biocompatibility as a delivery carrier. These results were further supported by the flow cytometry analysis, which revealed a significant decrease in CD86-positive cells and an increase in CD206-positive cells in the Treg-Exo and Treg-Exo-IKVAV groups (Fig. 3d and e). Specifically, the flow cytometric results showed that CD86-positive cell proportions decreased to 11.6% (Treg-Exo) and 11.1% (Treg-Exo-IKVAV) versus 40.8% (LPS), whereas CD206-positive cells increased to 23.5% (Treg-Exo) and 24.0% (Treg-Exo-IKVAV) from 3.91% (LPS) and 5.52% (Control), respectively.

To further elucidating how different treatments affected macrophage polarization, the RT-PCR was used to detect the gene expression of macrophages after different treatments for 4 days. As shown in Fig. 3f, the results revealed significant downregulation of classical M1 marker (iNOS) and pro-inflammatory cytokines (iNOS, TNF-α, and IL-1β) in both Treg-Exo and Treg-Exo-IKVAV cohorts compared to LPS-treated group. This pattern correlated with enhanced expression of alternative M2 polarization markers (CD206) and anti-inflammatory cytokine (IL-10), suggesting substantial phenotypic reprogramming towards anti-inflammatory macrophage subsets. To complement the transcriptional profiling, cytokine quantification was performed using enzyme-linked immunosorbent assays (ELISA). The Treg-Exo- and Treg-Exo-IKVAV-treated groups demonstrated suppressed production of proinflammatory cytokines (TNF-α, and IL-1β) alongside elevated anti-inflammatory IL-4 and IL-10 levels after 4 days of incubation (Figure S2a). The dual-regulatory effect on M1/M2 polarization reveals Treg-Exo-IKVAV’s distinctive mechanism for reprogramming macrophage polarization balance, simultaneously enhancing anti-inflammatory conditions while supporting tissue regeneration and repair.

Additionally, subsequent to complete phenotypic characterization, we applied RNA sequencing to decipher the molecular foundations of these observations. Comprehensive analysis of the 24-hour LPS-stimulated inflammatory system, incorporating KEGG pathway enrichment, methodically identified key signaling cascades controlling macrophage phenotypic conversion. As shown in Fig. 3g, prominent changes were detected in critical inflammation-related pathways, particularly Toll-like receptor, MAPK, NF-κB, and PI3K-Akt signaling pathways. GO analysis showed that cytokines activity and cytokine-mediated signaling pathway demonstrated strong association with inflammatory disease progression (Fig. 3h). Therefore, we speculate that Treg-Exo-IKVAV may modulates macrophage polarization balance by specifically regulating these signaling pathways, and some cytokines, such as IL-10, TGF-β, may also mediate Treg-Exo-IKVAV’s immunomodulatory effects on macrophage polarization transition (Fig. 3i).

While genomic analyses indicate that Treg-Exo-IKVAV modulates macrophage immune function through MAPK and PI3K-AKT signaling pathways, it is important to note that these pathways are primarily regulated at the protein level rather than by RNA. Therefore, a Western blot validation of key pathway components (e.g., phosphorylated forms of ERK, AKT) was performed to provide a more comprehensive understanding of the regulatory mechanism. As shown in Figure S2b, Treg-Exo-IKVAV treatment downregulated the phosphorylation levels of p38 and p65, but upregulated the phosphorylation level of AKT, further indicating that Treg-Exo-IKVAV modulates macrophage immune function through MAPK, NF-κB, and PI3K-AKT signaling pathways, consistent with the genomic analyses. Additionally, the corresponding uncropped original blot images were shown in Figure S2c.

Collectively, these data indicate that Treg-Exo-IKVAV exhibit potent immunomodulatory activity through dual mechanisms: suppression of pro-inflammatory signaling pathways and promotion of M2 macrophage differentiation. This immunoregulatory effects may create a pro-regenerative immune microenvironment for tissue regeneration.

Direct and immunoregulatory effects of Treg-Exo-IKVAV on NSCs differentiation in vitro

In the central nervous system, oligodendrocytes secrete myelin lipids (e.g., myelin basic protein, myelin-associated glycoprotein) to ensheath axons and accelerate electrical signal conduction [36]. They also maintain neuronal activity by supplying energy molecules such as lactate and glucose [36]. SCI induces apoptosis of oligodendrocytes near the lesion site via ischemia, calcium overload, and other mechanisms, leading to demyelination of axons [9]. Surviving oligodendrocytes or their precursor cells (oligodendrocyte precursor cells, OPCs) proliferate and migrate to the injured area to attempt repair. However, some oligodendrocytes may aberrantly myelinate non-myelinated regions (e.g., gray matter), disrupting neural circuitry [37]. Regulating the differentiation of OPCs into mature oligodendrocytes promotes axonal regeneration and remyelination [38]. Therefore, the effect of Treg-Exo-IKVAV on OPCs differentiation was studied. OLIG2 (Oligodendrocyte Transcription Factor 2) is a transcription factor specific to the oligodendrocyte lineage, remaining expressed throughout differentiation, whereas CC1 (Adenylate Cyclase-Activating Protein 1) serves as a marker for mature oligodendrocytes [38]. Therefore, immunostaining of OLIG2 (red) and CC1 (green) was performed to determine the effect of Treg-Exo-IKVAV on OPCs differentiation [38]. As shown in Figure S3a-3b, the Treg-Exo-IKVAV group showed significantly higher fluorescence intensity of green signal (CC1) in comparison to the control and Treg-Exo groups, indicating that Treg-Exo-IKVAV promoted OPCs differentiation into mature oligodendrocytes and may has a potent for axonal regeneration after SCI.

NSCs are undifferentiated cells in the central nervous system that possess both self-renewal capacity and multipotential differentiation capability [39]. After SCI, NSCs tend to aberrantly differentiate into glial cells, particularly astrocytes, rather than functional neurons within injured microenvironments. A previous study has shown that the proportion of NSC-derived glial cells in damaged spinal cord regions can reach 70–80% (compared to 10–15% during normal development), and this gliosis process forms dense glial scars that physically impede neural and axonal regeneration [40]. Recently, stem cell transplantation was reported to improve motor function of SCI mice [41]. However, stem cell-based therapies remain constrained by key limitations including immunogenicity and tumorigenicity. In comparison, recruiting endogenous NSCs at lesion sites and correcting abnormal differentiation to facilitate neuronal regeneration presents a more viable therapeutic strategy. Enlightened by these, the direct and immunomodulatory effects of Treg-Exo-IKVAV on NSCs differentiation were subsequently investigated.

First, the direct effect of Treg-Exo-IKVAV on NSCs differentiation was evaluated by co-culturing with NSCs isolated from postnatal rat (Fig. 4a). The MAP2 (green) and GFAP (red) immunostaining was performed, where MAP2 is a somatodendritic marker and GFAP serves as an astrocytic marker [42]. As shown in Fig. 4b, Treg-Exo-IKVAV significantly downregulated the GFAP expression and sharply increased density and brightness of green fluorescence signals compared with the Treg-Exo and control groups. The quantification shown in Fig. 4c further confirmed the observation. However, no significant difference was observed between the Treg-Exo and control groups in the intensity of MAP2 and GFAP signals. These evidences indicate that IKVAV peptide motifs endows Treg-Exo with the neuro-regenerative property, and Treg-Exo-IKVAV promoted neuronal differentiation of NSCs.

Fig. 4.

Treg-Exo-IKVAV promoted neuronal differentiation of NSCs in vitro.(a) Schematic diagram of the experiment evaluating the direct effect on NSCs differentiation. (b) Representative images of MAP2 (green) and GFAP (red) immunostaining. (c) Quantification of MAP2 and GFAP immunostaining of (b). (d) Schematic diagram of the experiment evaluating the immunoregulatory effect on NSCs differentiation. (e) Representative images of MAP2 (green) and GFAP (red) immunostaining. (f) Quantification of MAP2 and GFAP immunostaining of (e). (g) Neuronal differentiation related gene expression determined by RT-PCR. (h) KEGG analysis. (i) GO analysis. (j) Schematic illustration of the immunoregulatory mechanism of Treg-Exo-IKVAV on NSCs differentiation. Data are expressed as mean ± SD. Statistical significance was tested by one-way ANOVA with Tukey’s multiple comparison test (c, f, g). **p < 0.01; ***p < 0.001

To investigate the Treg-Exo-IKVAV’s immunoregulatory effect on NSCs differentiation, we designed a double-chamber co-culture system, allowing interactions between NSCs and macrophages (Fig. 4d). After co-culturing, MAP2/GFAP immunostaining were performed. As shown in Fig. 4e, The Treg-Exo and Treg-Exo-IKVAV nanovesicles effective increased MAP2 signal intensity and decreased GFAP intensity compared with control group. The quantitative analysis (Fig. 4f) further verified the observation. Additionally, the gene expression of MAP2, TUJ1, NeuN and GFAP was detected (Fig. 4g). Both Treg-Exo and Treg-Exo-IKVAV treatments significantly upregulated the expression of MAP2, TUJ1, and NeuN, while downregulating GFAP expression. The results indicate that both Treg-Exo and Treg-Exo-IKVAV have immunomodulatory ability, through which indirectly promoted neuronal differentiation of NSCs. The immunomodulatory ability of Treg-Exo and Treg-Exo-IKVAV nanovesicles may be due to active biomolecules inside Treg-Exo inhered from Tregs.

To delineate the molecular mechanisms underlying the Treg-Exo-IKVAV’s neuro-regenerative property, an RNA sequencing was performed. Whole-transcriptome analysis of the Treg-Exo-IKVAV-treated NSCs, integrated with KEGG pathway enrichment and GO analysis, systematically revealed critical signaling networks through which Treg-Exo-IKVAV promoted neuronal differentiation of NSCs. As shown in Fig. 4h, prominent changes were detected in JAK-STAT, NF-κB, signaling pathways, cytokine-cytokine receptor interaction, and cell adhesion molecules. And the GO analysis (Fig. 4i) of molecular function revealed significant changes in integrin binding, small molecule binding, and cytokine activity. Based on these results, we speculate that Treg-Exo-IKVAV can bind to cell surface integrins (such as α6β1), activate focal adhesion kinase (FAK) and the downstream Ras/MAPK (ERK1/2) or NF-κB pathway, thereby promoting the expression of neuronal differentiation markers (such as MAP2), consistent with previous studies [43]. In another hand, Treg-Exo-IKVAV may regulate JAK/STAT pathway and thereby downregulating GFAP expression (Fig. 4j).

However, the critical pathways involved in neural differentiation and regeneration, such as JAK/STAT and ERK1/2, are regulated through phosphorylation events. Therefore, a Western blot analyses of key components (e.g., p-STAT3, p-ERK1/2) was further performed to provide direct biochemical evidence of pathway activation. As shown in Figure S3c, increased phosphorylation levels of ERK1/2 and STAT3 were observed after Treg-Exo-IKVAV treatment, further verifying the genomic analyses. The corresponding uncropped original blot images were shown in Figure S3d.

Treg-Exo-IKVAV targeted inflammation

Enlightened by the in vitro immunomodulatory and neuro-regenerative properties, we advanced our research to explore the therapeutic effect on SCI mouse. Figure 5a depicted the methodology for the in vivo experiments. A weight-drop injury model was employed to induce SCI at the thoracic level (T10) in mouse. Then, caudal intravenous injection of PBS, Treg-Exo, and Treg-Exo-IKVAV was performed immediately.

Fig. 5.

Treg-Exo-IKVAV targeted inflammation and promoted motor functional behavior recovery of SCI mice. (a) Schematic illustration of the in vivo experiments. (b) Typical fluorescence images of ICG-labeled Treg-Exo and Treg-Exo-IKVAV. (c) Quantitative analysis of (b) (n = 3). (d) Ex vivo ICG signals of spinal cords harvested from SCI mice at 7. days after injection with Treg-Exo and Treg-Exo-IKVAV. (e) Ex vivo ICG signals of major organs (heart, liver, spleen, lung and kidney) harvested at 7. days after injection with Treg-Exo and Treg-Exo-IKVAV. (f) Quantitative analysis of relative fluorescence intensity in spinal cords and major organs (n = 3). (g) Representative images of footprint assay on day 28 after Treg-Exo and Treg-Exo-IKVAV treatments, and quantitative analysis of toe dragging and stride length (n = 5). (h) Typical images of the swimming test on day 28. (i) Quantification of the swimming test using Louisville Swim Scale (n = 3). (j) BMS scores in the different groups (n = 5). (k) H&E staining of bladder tissues in different groups on day 28, and quantitative analysis of bladder volume and bladder weight (n = 3). Data are expressed as mean ± SD. Statistical significance was tested by one-way ANOVA (g, k) or two-way ANOVA with Tukey’s multiple comparison test (c, f, i, j). *p < 0.05; **p < 0.01; ***p < 0.001

We first examined the targeting and accumulation of different nanovesicles in the injured site. To facilitate in vivo tracing of the biodistribution of different nanovesicles in mice, indocyanine (ICG) was used to label the Treg-Exo, and Treg-Exo-IKVAV, and a living imaging system was employed to observe the ICG-labeled nanovesicles in vivo at 1-, 7-, 14-, and 28-days post-injection. As shown in Fig. 5b, both Treg-Exo and Treg-Exo-IKVAV accumulated at the injured site after administration. The fluorescence intensities of ICG signals gradually decreased, with fluorescence detectable even after 28 days (Fig. 5c). These results indicated that both Treg-Exo and Treg-Exo-IKVAV exhibited inflammation tropism ability, and IKVAV modification did not alter the targeting property of Treg-Exo. Previous studies have demonstrated that exosomes inherently exhibit cell-type-specific targeting properties derived from their cellular origin [44]. These membrane-bound vesicles, secreted through exocytosis pathways, display surface signatures comprising lipid raft structures and adhesion proteins that recognize complementary receptors on recipient cells [45]. For example, Chen et al. [28] reported that M2-macrophage derived exosomes targeted the inflammation site of SCI. Upon binding to target cells, exosomes undergo membrane fusion to deliver their cargo, a diverse array of functional molecules including proteins, lipids, and non-coding RNAs (e.g. microRNAs, siRNAs), thereby modulating downstream cellular signaling cascades and inducing phenotypic alterations [46]. Therefore, we speculated that the inflammation-targeting property of Treg-Exo-IKVAV may be inherited from Tregs.

Subsequently, the biodistribution of these nanovesicles in mice was studied. At the 7-day following injection, spinal cord tissues and major organs were collected for ex vivo fluorescence quantification. As shown in Fig. 5d, Significant ICG fluorescence signals were observed at the SCI site in both the Treg-Exo and Treg-Exo-IKVAV groups, further confirming their inflammation-targeting capability. Additionally, the ICG-fluorescence signals were detected in the livers and kidneys in the Treg-Exo and Treg-Exo-IKVAV groups, demonstrating that both nanovesicles were mainly metabolized by the livers and kidneys (Fig. 5e and f). The ICG-fluorescence signals in heart and lung tissues may be caused by the circulation of nanovesicles in the bloodstream following tail vein injection.

Taken together, these evidences indicate Treg-Exo-IKVAV nanovesicles can specifically repair SCI through targeted mechanisms, and they are non-toxic, safe, and green biomaterials for SCI repair.

Treg-Exo-IKVAV nanovesicles promoted motor functional recovery of SCI mouse

The motor functional recovery of SCI mouse in the different groups was then evaluated. First, footprint behavioral assays were performed, and the typical footprints images were shown in Fig. 5g. The control group exhibited pronounced hind limb motor dysfunction, manifested by distinct gait disturbances characterized by compensatory forelimb weight-bearing (blue ink-marked forelimbs) and pronounced hind limb dragging (red ink-marked limbs). Therapeutic interventions with Treg-Exo demonstrated better restoration of hind limb strength by day 28 post-injury compared with the control group, though residual gait impairments persisted in the form of limb dragging. The most significant functional improvement was observed in mice receiving Treg-Exo-IKVAV treatment, which demonstrated the most remarkable recovery of hind limb motor function and substantial alleviation of gait abnormalities by day 28. The further quantification showed a marked decrease in the ratio of toe dragging and a significant increase in stride length of SCI mouse after Treg-Exo-IKVAV treatment. The locomotor recovery in SCI mouse was also evaluated using a swimming test. The typical images were shown in Fig. 5h. The control group showed severe locomotor deficits manifested by marked hindlimb immobility, exhibiting extensive reliance on forelimb propulsion for forward movement alongside pronounced truncal instability. In comparison, Treg-Exo-treated group displayed partial functional recovery, characterized by occasional hindlimb movements and moderate improvements in body stability compared to the SCI group. Notably, the Treg-Exo-IKVAV-treated group exhibited superior hindlimb functionality and significantly reduced forelimb dependence relative to other groups. This functional improvement was further substantiated by Louisville Swim Scale evaluations (Fig. 5i), which confirmed enhanced motor recovery in Treg-Exo-IKVAV-treated mice throughout the 28-day observation period.

Furthermore, the Basso Mouse Scale (BMS) scoring system was employed to objectively assess locomotor recovery in experimental cohorts following SCI (Fig. 5j). Quantitative evaluations were conducted at serial time points (1, 3, 7, 14, 21, and 28 days after injury). The control group demonstrated minimal incremental gains in BMS scores over the observation period, while Treg-Exo-treated SCI mouse exhibited a significant increase in BMS scores on days 7, 14, 21 and 28 post-injuries. Notably, the Treg-Exo-IKVAV group achieved best locomotor recovery outcomes, displaying the highest BMS scores among all experimental groups. Additionally, urinary damages are common complications caused by SCI, and in turn, the alleviation of urinary damages serves as an important index for therapeutic success. Figure 5k showed the typical H&E images of bladders in the different groups. Quantification demonstrated that Treg-Exo-IKVAV treatment significantly decreased the bladder weight and volumes.

Taken together, the cumulative evidence from these functional assessments confirms the curative potential of Treg-Exo-IKVAV nanovesicles in ameliorating locomotor deficits induced by traumatic SCI.

Treg-Exo-IKVAV alleviated macrophage infiltration and inflammation in vivo

Inflammatory response is one of the core pathological processes of SCI. After SCI, damaged cells release excessive proinflammatory cytokines such as TNF-α, IL-1β, and IL-6, which can recruit and activate inflammatory cells such as neutrophils and macrophages to exacerbate inflammation response [47]. As negative regulators of the immune system, Treg cells are theoretically capable of modulating the intensity and duration of immune responses following SCI, thereby preventing excessive immune activation [13]. However, the high concentration of inflammatory cytokines after SCI inhibited Tregs proliferation and impaired their function as a critical negative regulator of inflammation, thereby resulting in immune dysregulation. Therefore, it’s of great significance to study the in vivo immunoregulatory effects of Treg-Exo-IKVAV nanovesicles on neuroinflammation secondary to SCI.

First, MRI, a crucial tool for the assessment of the severity of SCI in clinical practices, was used to evaluate the recovery of spinal cord. The typical T2-MRI images captured at day 7 post injury was shown in Fig. 6a. Prominent high-signal regions were detected inside spinal cord in the control group. However, both Treg-Exo and Tre-Exo-IKVAV groups exhibited minimal high-signal intensity, suggesting lower inflammatory response and edema compared to the SCI group. Then, CD68 (a marker of macrophages) immunostaining was performed to assess macrophage infiltration at the injured sites in the different groups. Figure 6b revealed extensive accumulation of macrophages at the lesion site in the control group. Therapeutic administration of both Treg-Exo and Treg-Exo-IKVAV nanovesicles demonstrated marked diminishment in both cellular population density and spatial coverage compared to the control group, suggesting potent anti-inflammatory efficacy of Treg-Exo and Treg-Exo-IKVAV. However, no significant difference was found between Treg-Exo and Treg-Exo-IKVAV groups.

Fig. 6.

Treg-Exo-IKVAV attenuated macrophage infiltration and inhibited macrophage-induced inflammation response secondary to SCI.(a) Sagittal T2-MRI images on day 7. (b) Immunostaining of CD68 in different groups on day 7. (c) Quantification of CD68 immunostaining (n = 3). (d) Co-immunostaining of CD68 (green) and iNOS (red), and CD68 (green) and Arg1 (red) in the different groups. (e-f) Quantification of CD68, iNOS, and Arg1 immunofluorescence staining (n = 3). (g) RT-PCR detection of the gene expression (iNOS, TNF-α, IL-1β, IL-10, and Arg1) (n = 3). (h) Levels of TNF-α, IL-6, CXCL10, IL-4 and IL-10 detected by ELISA (n = 5). Data are presented as mean ± SD. Statistical significance was tested by one-way ANOVA (c, e, f, g, h) with Tukey’s multiple comparison test. *p < 0.05; **p < 0.01; ***p < 0.001

Subsequently, to delineate macrophage polarization dynamics following therapeutic interventions, we employed iNOS and Arg1 as discriminatory biomarkers for M1 (pro-inflammatory) and M2 (anti-inflammatory) phenotypes. Immuno-histological staining revealed predominant M1 polarization in the control group, characterized by abundant iNOS⁺ immunoreactivity and scarce Arg1⁺ expression (Fig. 6d). Both Treg-Exo and Treg-Exo-IKVAV treatments accelerated the phenotypic conversion from inflammatory M1 to regenerative M2 states in macrophages, as demonstrated by the reduced iNOS⁺/CD68⁺ macrophage proportion and the increased Arg1⁺/CD68⁺ population (Fig. 6e and f). This polarization shift was further substantiated by RT-PCR (Fig. 6g). RT-qPCR analysis at 7 days post-injury showed elevated M2-associated gene transcripts (Arg1, IL-10) alongside suppressed M1-related expression (TNF-α, iNOS, IL-1β) in both treated groups compared to the control group. Notably, comparable efficacy was observed between Treg-Exo and Treg-Exo-IKVAV treatments regarding macrophage reprogramming, with no statistically significant inter-group differences detected. Additionally, similar results were observed in ELISA detection which was shown in Fig. 6h.

Taken together, the results verify the Treg-Exo and Treg-Exo-IKVAV’s immunomodulatory effect in vivo. They inhibited macrophage infiltration and promoted macrophage reprogram, thereby creating a pro-regenerative immune microenvironment and promoting motor functional recovery of SCI mice.

Treg-Exo-IKVAV nanovesicles promoted neuronal differentiation of NSCs in vivo

Enlightened by the in vitro neuro-regenerative property, the effects of Treg-Exo-IKVAV nanovesicle on NSCs differentiation in vivo was further investigated. First, the NeuN immunostaining, a marker for mature neurons, was performed. Figure 7a displays fluorescent NeuN⁺ cell distributions across specified zones (Z1-Z4) after various treatments. The Treg-Exo-treated group demonstrated enhanced NeuN⁺ neuronal counts relative to controls. Notably, Comparative analysis across all four regions revealed highest NeuN⁺ cell density in the Treg-Exo-IKVAV group compared to the other treatments (Fig. 7b). These observations suggested that Treg-Exo-IKVAV nanovesicles promoted neuronal differentiation of NSCs. Figure 7c presents the typical TUJ1 immunostaining images, and similar results were observed. The further quantitative evaluation (Fig. 7d) indicates the Treg-Exo-IKVAV group achieved approximately 35.4% TUJ1⁺ cell prevalence, nearly quadruple the control level, substantiating enhanced neuronal differentiation of NSCs. Notably, Treg-Exo exhibited more pronounced in vivo neural regenerative effects compared to its in vitro performance, potentially attributed to immunomodulatory properties, alleviating neuroinflammation and thereby facilitating neuronal differentiation of NSCs.

Fig. 7.

Treg-Exo-IKVAV promoted neuronal differentiation of NSCs, and axonal regeneration and remyelination.(a) Immunofluorescence staining of NeuN to label the mature neurons on day 28. (b) Quantification of NeuN⁺ cells in Z1-Z4 regions adjacent to the lesion core (n = 3). (c) Immunostaining of TUJ1 to label the newly born neurons on day 28. (d) Percent of TUJ1⁺/DAPI⁺ cells (n = 5). (e) Quantification analysis of (f) co-immunostaining of NF-200 (green) and GFAP (red), and (g) co-immunostaining of MBP (green) and GFAP (red) in the different groups at 28 days after SCI. Data are expressed as mean ± SD. Statistical significance was tested by one-way ANOVA (d) or two-way ANOVA with Tukey’s multiple comparison test (c). *p < 0.05; **p < 0.01; ***p < 0.001

Treg-Exo-IKVAV nanovesicles promoted axonal regeneration and remyelination in vivo

Axonal regeneration is a central component of SCI repair and a critical limiting factor for functional recovery [48]. Neuronal functionality critically depends on the proper axonal connection. Enlightened by the in vitro results that Treg-Exo-IKVAV promoted OPCs differentiation into mature oligodendrocytes, the effect of Treg-Exo-IKVAV on axonal regeneration was further evaluated in vivo.

First, NF-200 (neurofilament marker) and GFAP (glial scar indicator) immunostaining was performed. The typical images were shown in Fig. 7f. The control group exhibited pronounced GFAP-positive gliosis, whereas magnified imaging revealed sparse green fluorescence intensity. Post-Treg-Exo treatment showed modest fluorescence enhancement in NF-200 signals alongside GFAP signal attenuation. This may potentially be attributed to immunomodulatory properties of Treg-Exo, alleviating neuroinflammation and thereby creating a favorable immune microenvironment for axonal regeneration. Notably, Treg-Exo-IKVAV administration inhibited astrocyte formation and robust upregulation of NF-200 expression at the lesion site (Fig. 7e). Mechanistically, this regenerative response may originate from the IKVAV peptide motifs engineered into the surface of Treg-Exo. Previous studies have demonstrated that the IKVAV peptide binds to integrins (e.g., α6B1), activates FAK and the downstream Ras/ERK pathway, and promotes the differentiation of neural precursor cells into neurons by upregulating markers such as NeuroD and MAP2 [43]. Additionally, the inherent anti-inflammatory properties of Treg-Exo-IKVAV could mitigate post-injury inflammatory cytokine release, thereby overcoming microenvironmental barriers to axonal regeneration.

The depletion of oligodendrocytes following traumatic SCI disrupts axonal myelination, resulting in severe neurological deficits [17]. Existing evidence has established a strong association between endogenous myelin restoration and functional recovery in SCI models [49]. In this study, to evaluate the effect of Treg-Exo-IKVAV nanovesicles on axonal remyelination, MBP/GFAP (green/red) immunostaining was conducted. Similar to observations in NF-200 staining, minimal green fluorescence intensity was observed in the SCI group (Fig. 7g). In contrast, Treg-Exo-IKVAV treatment elicited significant increases in both the density and brightness of green fluorescence signals compared to Treg-Exo and control groups (Fig. 7e). These findings indicated that Treg-Exo-IKVAV effectively facilitated axonal regeneration and remyelination post-SCI, highlighting therapeutic potential for SCI repair.

Collectively, according to the in vitro and in vivo results, the nanovesicle system is rationally engineered to initiate early immunosuppression via Treg-derived exosomal contents, followed by delayed yet spatially targeted neuroregeneration through IKVAV-mediated signaling. This design enables spatiotemporally sequential engagement with the neuroimmune microenvironment, providing a synergistic platform for SCI repair.

Biosafety evaluation

To evaluate the in vitro cytotoxicity of Treg-Exo-IKVAV nanovesicles, the live/dead staining was performed after co-culturing with macrophages for 24 h. The typical images were shown in Fig. 8a. No significant difference was observed among the groups. Furthermore, CCK-8 was performed after co-culturing with NSCs (Fig. 8b). Treg-Exo and Treg-Exo-IKVAV significantly improved cell viability of NSCs. These evidences indicate that Treg-Exo-IKVAV has good biocompatibility in vitro.

Fig. 8.

In vitro and in vivo biosafety of Treg-Exo-IKVAV nanovesicles.(a) Typical images of Live/Dead staining. (b) Viability of NSCs after cultured with Treg-Exo or Treg-Exo-IKVAV for 24 h determined by CCK-8 (n = 5). (c) H&E staining of heart, liver, spleen, lung, and kidney tissues in the different groups. (d) Biochemical markers relevant to hepatic (ALT, AST) and kidney (UA, BUN) function (n = 5). Data are expressed as mean ± SD. Statistical significance was tested by one-way ANOVA (b, d) with Tukey’s multiple comparison test. ***p < 0.001

To evaluate the in vivo biosafety of Treg-Exo-IKVAV, major organs including the heart, liver, spleen, lung, and kidney from mice in different groups at 28 days were harvested for histological analysis. As demonstrated in Fig. 8c, no noticeable damage or lesions were detected in the different groups under light microscopic examination. Besides, blood test for biochemical markers relevant to hepatic (ALT, alanine transaminase; AST, aspartate transaminase) and kidney (UA, uric acid; BUN, blood urea nitrogen) function showed no systemic toxicity in all groups (Fig. 8d). These evidences indicate that Treg-Exo-IKVAV nanovesicles showed excellent biocompatibility in vivo.

Conclusions

The present study developed a novel Treg-derived exosome-peptide composite (Treg-Exo-IKVAV) by covalently conjugating the laminin mimetic peptide IKVAV to the surface of Treg exosomes through click chemistry. This innovative design not only preserves the natural immunomodulatory properties of exosomes but also endows them with targeted neuro-regenerative capability, achieving a synergistic effect between immunomodulation and neuronal regeneration. Experiment results demonstrate that Treg-Exo-IKVAV effectively induces macrophage polarization toward the M2 phenotype, significantly improving the local immune microenvironment in injured site and reducing inflammatory responses. Meanwhile, the composite promotes neuronal differentiation of NSCs directly and by immunomodulation. In SCI mouse model, Treg-Exo-IKVAV demonstrates remarkable therapeutic efficacy, with significantly better performance in promoting neuronal regeneration, axonal regeneration and remyelination.

These results not only confirm the effectiveness of Treg-Exo-IKVAV as a novel SCI repair material but also reveal the close relationship between immune regulation and neuronal regeneration, providing important theoretical basis for developing immune-based SCI repair strategies. The “exosome-peptide” composite design proposed in this study offers a new research direction in tissue engineering and regenerative medicine, with significant scientific implications and clinical application prospects.

However, this study has certain limitations. First, even though the click chemistry is an efficient, low-cost, and safe method to conjugate IKVAV onto surfaces of Treg-Exo, and the modification efficiency is relatively high, we found that not all exosomes can be successfully modified, and the modification efficiency was approximately 90%. Prior to conducting in vitro and in vivo experiments, Treg-Exo-IKVAV was subjected to additional isolation and purification to meet the experimental requirements. Nonetheless, further study is required to develop a more efficient modification method in the future. Second, future studies should further investigate the therapeutic efficacy of Treg-Exo-IKVAV in SCI models across additional species to advance clinical translation. Third, the biological activity of IKVAV is highly dependent on its mode of presentation. Soluble, unconjugated IKVAV peptides are prone to rapid degradation and diffusion, and lack the multivalent spatial configuration necessary for effective receptor engagement. Therefore, in this study, we did not set up a separate IKVAV group.

Materials and methods

Reagent and chemicals

The peptides Acryl-Gly-Ile-Lys-Val-Ala-Val (IKVAV, molecular weight [MW] = 640.3 g/mol) and azido-IKVAV were chemically synthesized via NJPeptide (Nanjing, China). The crosslinker DBCO-sulfo-N-hydroxysuccinimidyl ester (DBCO-sulfo-NHS) was purchased from Sigma-Aldrich (USA), while the fluorophores DiO were purchased from Beyotime (China). The green-fluorescent dye FITC was obtained from Biosharp (China).

Preparation of Treg-Exo and Treg-Exo-IKVAV

Isolation of Treg-Exo: Tregs were isolated from spleens using a standard Ficoll-Paque gradient centrifugation protocol [25]. Briefly, CD4 + CD25 + Foxp3 + Treg cells from mouse spleen were isolated and cultured in RPMI-1640 supplemented with 10% FBS, 2 mM L-glutamine, 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 °C in 5% CO₂ for 48 h. Then, cell supernatant was collected and exosomes were isolated through differential centrifugation: 300 × g (15 min) for cell removal, 10,000 × g (15 min) for debris clearance, and final precipitation at 100,000 × g (3 h) using Beckman Optima XE-100. Pelleted Treg-Exo were resuspended in PBS and underwent repeated ultracentrifugation. Further purify via iodixanol density gradient centrifugation (6–18%), collect fractions at 1.10–1.14 g/mL, and dialyze against PBS to remove iodixanol.

Preparation of Treg-Exo-IKVAV: To conjugate exosomes with IKVAV, a two-step reaction was proposed. First, Treg-Exo (0.5 mg mL − 1 in PBS) were reacted with 3 mM DBCO-sulfo-NHS for 4 h at 20 °C with constant rotation, followed by ultrafiltration to remove unreacted reagents. Copper-free click chemistry was then performed by incubating DBCO-modified exosomes with 0.5 µM azido-IKVAV (12 h, 4 °C, rotating mixer). Post-reaction purification involved: (1) sucrose cushion centrifugation (30% in D2O, 160,000 × g, 1 h) to eliminate free peptides, and (2) PBS washing before final storage at − 80 °C.

Characterization of Treg-Exo and Treg-Exo-IKVAV

The morphological characteristics of Treg-Exo and Treg-Exo-IKVAV nanovesicles were visualized using transmission electron microscopy (TEM; Tecnai Spirit, FEI, USA). Zeta potential and particle size distributions of exosomes were determined via nanoparticle tracking analysis (NTA; ZetaView PMX 110, Particle Metrix, Germany). Dynamic light scattering (DLS; Malvern Zetasizer Nano ZS90, Malvern Instruments, UK) was employed to quantify polydispersity index (PDI) values for both Treg-Exo and Treg-Exo-IKVAV. Western blot analysis and flow cytometry were conducted to validate exosomal biomarker expression (CD9, TSG101, CD63 and CD81). For fluorescent labeling, azido-IKVAV conjugates bearing Cy5.5 fluorophore were prepared through NJPeptide’s synthesis platform (Nanjing, China). Treg-Exo samples were stained with DiO (Beyotime, China) and subsequently analyzed by confocal laser scanning microscopy (CLSM; TCS SP8, Leica Microsystems, Germany) to assess fluorescent labeling efficiency.

Cells culture of Raw 264.7

RAW264.7 macrophages were obtained from Procell Life Science & Technology (Wuhan, China) and maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (Life Technologies, USA) at 37 °C in 5% CO₂. Cells were routinely passaged every 3–4 days to maintain exponential growth.

Cellular uptake experiment

To investigate RAW 264.7 macrophage uptake of Treg-Exo and Treg-Exo-IKVAV, cells were incubated with either compound for 24 h followed by fixation with 4% paraformaldehyde (PFA). Post-fixation, Treg-Exo and cells were labeled with FITC (green fluorescence) for Treg-Exo and Treg-Exo-IKVAV visualization and DAPI (blue) for nuclear staining, respectively. Fluorescent signals were visualized using a Leica upright microscope (Germany).

Detection of macrophage phenotype transition

RAW264.7 macrophages were cultivated in a 24-well plate for 6 h, followed by treatment with 100 ng/mL lipopolysaccharide (LPS) for 24 h to induce M1 polarization. Subsequently, Treg-Exo and Treg-Exo-IKVAV (2.0 × 10¹¹ particles/mL) were introduced and incubated for an additional 24 h. Macrophage phenotype transitions were comprehensively assessed through qRT-PCR, immunofluorescence staining, flow cytometry, and ELISA. For qRT-PCR analysis, mRNA expression levels of M1-associated genes (iNOS, TNF-α, IL-1β) and M2-associated genes (CD206, IL-10) were quantified using the 2 − ΔΔCt method, as detailed previously. Primer sequences are listed in Table S1 of the Supporting Information. Immunofluorescence staining was performed to detect CD206 (M2 marker) and iNOS (M1 marker), following established protocols. For flow cytometric analysis, cells expressing CD86 and lacking CD206 were classified as M1 macrophages, while CD206-positive/CD86-negative cells were designated as M2 macrophages. Post-staining, cells were resuspended in Flow cytometry buffer and analyzed using a Cytoflex ADP instrument (Beckman Coulter, USA). Data were processed with FlowJo v10.0.7 software. The cell supernatant was collected for ELISA detection of the levels of inflammatory cytokines (TNF-α, IL-1β, IL-4, and IL-10) following the manufacturer’s instructions.

RNA sequencing

To investigate the immunomodulatory and neuronal regenerative effects of Treg-Exo-IKVAV, we performed whole-genome transcriptomic profiling in RAW 264.7 macrophages or NSCs, respectively. Cells were plated at 5 × 10⁵ cells/well in 6-well plates and co-treated with 100 ng/mL ultrapure LPS (InvivoGen, tlrl-pelps) and/or Treg-Exo-IKVAV for 24–72 h hours under standard conditions. Total RNA was extracted using TRIzol™ according to the manufacturer’s protocol, and RNA integrity was assessed and verified using the Bioanalyzer 2100. Strand-specific RNA sequencing libraries were prepared using the NEBNext^® Ultra™ II Directional RNA Library Prep Kit (E7760L, NEB) with poly-A selection, and sequencing was performed on the Illumina NovaSeq 6000 platform (150 bp paired-end reads, 50 million reads per sample) by Novogene Co., Ltd. (Shanghai). Raw data processing included adapter trimming (Trimmomatic v0.39), quality control (FastQC v0.11.9), and alignment to the mm10 reference genome (STAR v2.7.10a). Differential gene expression analysis was conducted using DESeq2 (v1.38.3) with thresholds set at | log2(fold change) | > 1 and adjusted p-value < 0.05. Functional enrichment analysis of significantly regulated genes was performed through Gene Ontology (GO) and KEGG pathway analysis (clusterProfiler v4.8.1), and heatmap visualization was generated using ComplexHeatmap (v2.16.0).

Isolation and culture of oligodendrocyte precursor cells (OPCs), and evaluation of OPCs differentiation

The oligodendrocyte precursor cells (OPCs) were isolated from SD rats following institutional animal care guidelines. Brains were rapidly dissected under ice-cold PBS, and the corpus callosum was minced into 1 mm³ fragments using a sterile razor blade. Tissue digestion was performed sequentially with 0.1% collagenase IV (Worthington Biochemicals) at 37 °C for 45 min and 0.25% trypsin-EDTA (Gibco) at 37 °C for 20 min. Enzymatic reaction was terminated by adding ice-cold FBS (10% v/v), followed by mechanical dissociation through 100 μm nylon strainers. Cell suspension was centrifuged at 800×g for 10 min to remove debris, then resuspended in proliferation medium containing DMEM/F-12 (Gibco) supplemented with 2% FBS, 10 ng/mL epidermal growth factor (EGF; Peprotech), 10 ng/mL fibroblast growth factor 2 (FGF-2; Peprotech), and 1% penicillin-streptomycin. OPC enrichment was achieved via density gradient centrifugation using 70% Percoll (Sigma-Aldrich), collecting the interface layer corresponding to 10–20% gradient. To evaluate the effect of Treg-Exo-IKVAV on OPCs differentiation, the OPCs were co-cultured with different samples for 7 days. Subsequently, the OLIG2 and CC1 immunostaining was performed, and the typical images were captured by CLSM.

NSCs isolation and culture

Neural stem cells (NSCs) were derived from the hippocampal regions of embryonic Sprague-Dawley rats. Tissues were enzymatically dissociated using Accutase^® solution (Gibco, USA) at 37 °C for 20 min, followed by mechanical dispersion under sterile conditions to generate single-cell suspensions. Cellular aggregates were removed through a 200-µm mesh filter. Purified cells were cultured as neurospheres in T15 flasks (Corning, USA) with DMEM/F-12 medium (Gibco, USA) supplemented with 2% B27 supplement (Gibco, USA), 20 ng/mL epidermal growth factor (EGF; PeproTech, USA), 20 ng/mL basic fibroblast growth factor (bFGF; PeproTech, USA), and 1% penicillin-streptomycin (Invitrogen, USA) at 37 °C in 5% CO₂. Cells between passages 2–4 were utilized for subsequent experiments.

Evaluation of Treg-Exo-IKVAV’s direct effect on NSCs differentiation

To initiate NSCs differentiation assays, neurospheres were enzymatically dissociated into single cells using Accutase^® (Gibco, China) and seeded at a density of 1 × 10⁴ cells/well in a 24-well plate. Cells were treated with Treg-Exo or Treg-Exo-IKVAV (2.0 × 10¹¹ Particles/mL, 50 µL) for 7 days to induce differentiation. Phenotypic changes were evaluated through immunocytochemical staining using GFAP (CST, USA) and MAP2 (CST, USA) antibodies, which specifically mark astrocytes and neurons, respectively.

Evaluation of Treg-Exo-IKVAV’s Immunomodulatory effect on NSCs differentiation

To investigate the Treg-Exo-IKVAV’s immunoregulatory effect on NSCs differentiation, we designed a double-chamber co-culture system, allowing interactions between NSCs and macrophages. After co-culturing, MAP2/GFAP immunostaining were performed. We placed 0.4-µm-pore trans-well inserts on the 48-well plates and cultured the upper-layer NSCs. The macrophages at a density of 1.5 × 10⁵ cells per well were seeded, and Treg-Exo or Treg-Exo-IKVAV nanovesicles were added (2.0 × 10¹¹ Particles/mL, 50 µL) at the bottom of the well. Then, phenotypic changes were evaluated through immunocytochemical staining using GFAP (CST, USA) and MAP2 (CST, USA) antibodies, which specifically mark astrocytes and neurons, respectively. Additionally, mRNA expression levels of TUJ1, NeuN, GFAP and MAP2 were quantified via RT-PCR. Primer sequences are listed in Table S1 of the Supporting Information.

SCI animal model

A contusive SCI model was induced in female C57BL/6 mice (8–10 weeks). Animals were randomly allocated into three experimental groups: (1) control, (2) Treg-Exo-treated, and (3) Treg-Exo-IKVAV-treated. The surgical protocol followed established procedures. Mice received intraperitoneal isoflurane anesthesia prior to a midline incision at T9-T11 vertebral levels. A laminectomy was performed at T10 to expose the spinal cord, after which a 5 g weight was dropped from 5 cm height onto the cord using a precision impactor (RWD, USA). Post-injury care included suture closure of the musculature and skin, prophylactic antibiotic administration, and twice-daily manual bladder expression until spontaneous voiding resumed. Nanoparticle treatments (2.0 × 10¹¹ Particles/mL, 50 µL) were delivered via tail vein injection on the day of injury, while controls received isotonic saline.

In vivo imaging

In vivo tracking was conducted using female C57BL/6 mice subjected to spinal cord contusion, which were randomly allocated into two cohorts: (1) Treg-Exo-ICG-treated and (2) Treg-Exo-IKVAV-ICG-treated groups. The near-infrared dye ICG (MCE, USA) was incorporated into both Treg-Exo and Treg-Exo-IKVAV nanoparticles. Equivalent doses of 50 µL Treg-Exo-ICG, and Treg-Exo-IKVAV-ICG suspensions (2.0 × 10¹¹ Particles/mL, 50 µL) were administered via tail vein injection immediately post-injury, respectively. Nanoparticle biodistribution was longitudinally tracked using the IVIS^® Lumina Series III imaging system (PerkinElmer, USA) with standardized imaging parameters of 785 nm excitation wavelength and 810 nm emission detection. Additionally, spinal cords and major organs (including heart, liver, spleen, lung and kidneys) were harvested at 7. days after SCI. Nanoparticle biodistribution in spinal cords and major organs were also detected.

Evaluation of motor functional recovery

Gait analysis was conducted 28 days post-surgical induction in SCI mice. Forelimbs were stained with blue ink, while hind paws were marked with red ink. Animals were motivated to traverse a clean white platform within their cage, enabling digital documentation of footprints. Quantitative gait parameters including stride length and toe dragging were measured through image analysis software. Motor recovery was assessed using validated behavioral scales: the Basso Mouse Scale (BMS) evaluated hindlimb movement coordination (0–9 points) during 5-minute open-field observation, while the Louisville Swim Score (LSS) quantified functional recovery through forelimb usage, hindlimb mobility, and body posture. All scoring assessments were performed independently by three blinded observers using standardized protocols.

Histological analysis

Animals were humanely terminated at 7-day and 28-day post-injury time points. Systemic perfusion via cardiac route was performed sequentially with phosphate-buffered saline and 4% formaldehyde solution. Spinal cords were harvested, postfix in 4% formaldehyde, paraffin-embedded, and sectioned into 20 μm-thick coronal slices. Immunofluorescent staining was carried out to quantify inflammatory response and regenerative processes. Macrophage populations were identified using CD68-specific antibody (Abcam, USA), with subsequent differentiation subtypes classified through iNOS (Abcam, USA) and Arg1 (Abcam, USA) immunolabeling. Additional markers including glial fibrillary acidic protein (GFAP; Boster, China), neurofilament 200 (NF200; Abcam, USA), and myelin basic protein (MBP; CST, USA) were employed to assess spinal cord tissue remodeling.

Biocompatibility of Treg-Exo-IKVAV

To evaluate the in vitro cytotoxicity of Treg-Exo-IKVAV nanovesicles, the live/dead staining was performed after co-culturing with macrophages for 24 h according to the manufacturer’s instructions. Furthermore, The cell viability of NSCs co-cultured with various samples (2.0 × 10¹¹ Particles/mL, 50 µL) for 24 h was determined by a CCK-8 (Dojindo, Japan) kit following the manufacturer’s instructions.

To evaluate the in vivo biosafety of Treg-Exo-IKVAV, major organs including the heart, liver, spleen, lung, and kidney from mice in different groups at 28 days were harvested for H&E staining. Besides, blood test for biochemical markers relevant to hepatic (ALT, alanine transaminase; AST, aspartate transaminase) and kidney (UA, uric acid; BUN, blood urea nitrogen) function were performed.

Statistical analysis

The data from experiments are presented as the mean ± standard deviation (SD). Sample sizes as well as p-values and their significance are indicated in figure captions. The IBM SPSS Statistics and GraphPad Prism were used for the statistical analysis. One-way analysis of variance (ANOVA) was performed to compare one factor in two or more groups. Two-way ANOVA was performed to compare two factors in two or more group. Tukey’s post-hoc multiple comparison test was used for significant main effect. A p value of less than 0.05 deemed to indicate statistical significance across all tests (*p < 0.05; **p < 0.01; ***p < 0.001.).

Author contributions

Su Xinjin: Investigation, Writing – original draft. Gu Changjiang: Data curation, Writing – original draft. Zeng Feihui: Data curation, Funding acquisition. Wei Ziheng: Conceptualization. Kong Qingjie: Data curation, Funding acquisition. Sun Weijin: Supervision. Zhu Chao: Funding acquisition, Conceptualization. Chen Xiongsheng: Writing – review & editing, Conceptualization Project administration.

Funding

declaration.

The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (No. 82272496), and the Science and Technology Innovation Joint Fund Project of Fujian Province (No. 2021Y9053), and the Shanghai Committee of Science and Technology (No. 23Y11903700).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Female C57BL/6 (8–10 weeks old) were obtained from Hangzhou Ziyuan Laboratory Animal Technology Co., Ltd. All care and treatment of experimental animals were in strict accordance with the guidelines of the Association of Assessment and Accreditation of Laboratory Animal Care approved by the Animal Care Committee at Renji Hospital, Shanghai Jiao Tong University School of Medicine (No. 2025229).

Consent for publication

All authors consent for publication.

Competing interests

The authors declare no competing interests.