Bioprinting of live platelet-loaded nerve conduit using energy-dissipative hydrogel

Abstract

Platelets show great promise for nerve repair due to their abundant release of growth factors. However, they often suffer from rapid activation-induced burst release of the cargo, making it challenging to develop long-acting platelet preparation. Here, we show a biosynthetic nerve conduit containing platelets with prolonged survival for peripheral nerve repair. This conduit was rapidly fabricated using a customized 3D printer by coating a platelet-loaded Pluronic F127 diacrylate hydrogel onto an electrospinning polycaprolactone conduit. The hydrogel can protect the platelets from stress-induced activation during deformation through its nanocolloid-based energy-dissipative centers, achieving a platelet survival rate of 34.7 % after 600 compression cycles. Platelets survived in this hydrogel for over 2 weeks, enabling the sustained release of bioactive cargo such as NGF and VEGF for more than 20 days. This conduit also had good mechanical properties, including compression and stretch resistance, to support surgical suturing and structural stability in vivo. Twelve weeks post-implantation, this conduit efficiently promoted nerve repair with functional outcomes by providing a growth factor-rich microenvironment, demonstrating potential clinical application.

Graphical abstract

Highlights

• •Long-acting platelets preparation for sustained bioactive growth factor release.
• •Energy-dissipative hydrogels efficiently prevented rapid platelet activation.
• •A customized 3D printer was used to fast fabricate the bio-synthetic nerve conduit.
• •The live platelet-loaded nerve conduit had excellent mechanical properties.
• •The conduit could efficiently repair the peripheral nerve defects.

1. Introduction

Peripheral nerve injuries, one of the most common traumatic injuries to the nervous system, impact millions of people worldwide [1]. Although the peripheral nervous system has intrinsic regenerative capabilities, spontaneous nerve repair often results in unsatisfactory functional recovery for nerve defects [2,3]. Autografts, the gold standard for critical nerve defect repair, are limited by donor site availability, morbidity, and the requirement of multiple surgeries [4,5]. Nerve guidance conduits derived from biomaterials have gained significant attention as potential alternatives, as they can provide physical support and guidance for nerve regeneration [6,7]. However, traditional nerve conduits often show limited therapeutic efficacy, frequently leading to suboptimal restoration of sensory and motor functions, which may be attributed to their insufficient bioactivity [8,9]. Previous studies have demonstrated that the introduction of physical, chemical, or bioactive components into traditional conduits can improve nerve regeneration [[10], [11], [12], [13]]. Among these, bioactive substances such as growth factors, extracellular matrix proteins, and cells can provide essential biological cues to actively support and guide axonal regeneration [14]. The incorporation of these functional elements into nerve conduits offers a promising approach to promote nerve repair [15,16].

Platelets show significant potential in the treatment of peripheral nerve defects due to their ability to release multiple growth factors, including nerve growth factor (NGF) and vascular endothelial growth factor (VEGF) [17,18]. Although direct injection of platelets at injury sites has exhibited therapeutic efficacy, its clinical application is limited by the burst release and degradation of cargo due to the rapid platelets activation [19]. To overcome these challenges, hydrogels have been developed as carriers for platelets to reduce the activation risk. In static environment, the 3D molecular networks of hydrogels can protect platelets from external stimuli as barriers [20]. These hydrogel networks not only allow the sustained release of growth factors but also protect them from enzymatic degradation, thereby maintaining their biological function and therapeutic efficacy [21]. Consequently, platelet-loaded hydrogel conduits exhibit considerable promise for treating peripheral nerve injuries under static conditions. Nevertheless, hydrogel conduits often suffer from deformation during the preparation and application process [22]. Such deformation under dynamic conditions often leads to localized stress concentration which would cause undesirable platelet activation [23]. Therefore, it is crucial to develop conduits that can effectively protect platelets from fast activation under both static and dynamic situations.

Hydrogels with energy-dissipation capabilities could mitigate stress, offering a promising approach to protect platelets under dynamic physiological conditions [[24], [25], [26]]. By evenly distributing mechanical forces, they minimize localized stress concentrations and lower the mechanical load. This mechanical regulation would reduce deformation-induced platelet activation, thereby preserving its integrity and biological function. Pluronic F127 diacrylate (F127DA) hydrogel with good biocompatibility has been widely used in biomedical fields. Notably, it features a covalently cross-linked micellar network, which is capable of dissipating energy through intermolecular friction and reversible reorganization within hydrophobic domains [26,27]. Thus, F127DA hydrogel might be a promising candidate to protect platelets in mechanically dynamic and static conditions.

3D bioprinting technology is an advanced tool for manufacturing nerve conduits with the advantages in fast and precisely fabricating customized structures [28]. This approach enables efficient incorporation of biomaterials and biofactors in a mild manner. In this work, we identified that Pluronic F127 diacrylate (F127DA) hydrogel can effectively protect platelets from activation through its unique energy-dissipative nanostructures. Furthermore, we fabricated 3D-printed conduits consisting of a stretch-resistant electrospun polycaprolactone inner layer and a compression-resistant energy-dissipative hydrogel outer layer containing platelets (Fig. 1). The in vivo study indicated that these conduits could accelerate nerve repair and functional recovery by creating a growth factor-rich regenerative microenvironment. This work would advance the design of novel conduits for effective peripheral nerve repair.

Fig. 1.

Schematic illustration of the platelet-loaded nerve conduits, in which the hydrogel could protect platelets from fast activation through its nanosized energy-dissipative centers, enabling the sustained release of cargo to promote nerve repair.

2. Results

2.1. Fast 3D printing of the platelets-loaded nerve conduits

The survival of platelets within the hydrogels was investigated under cyclic compressive loading up to their maximum strain (Fig. S1). Moreover, GelMA and PEGDA, representative photopolymerizable hydrogels widely used in 3D bioprinting that exhibited distinct mechanical and structural characteristics, were selected as controls: PEGDA served as a rigid, synthetic, and bioinert polymer [29], whereas GelMA as a bioactive, naturally derived hydrogel with relatively soft mechanical properties [30]. As shown in Fig. 2A, the activation rates of platelets within all hydrogels elevated with increased cycles of the applied stress at maximum strain. Notably, after 600 cycles of compression, the platelet survival rate of the F127DA group was 34.7 ± 4.0 %, significantly higher than that of the PEGDA group (7.1 ± 1.6 %, p < 0.05) and the GelMA group (13.1 ± 2.4 %, p < 0.05) (Fig. 2B).

Fig. 2.

The effect of energy-dissipative hydrogel on platelet survival. (A) Representative images showing the survival of platelets in platelet-loaded hydrogels after compression. Green fluorescence represents non-activated platelets. (B) Survival rates of platelets in hydrogels after compression (n = 3, technical replicates). (C) Statistical analysis of NGF released from the platelet-loaded hydrogels after compression (n = 3, technical replicates). (D) Platelet survival rates in hydrogels under static environment for 10 days (n = 3, technical replicates). (E) NGF release profiles from platelet-loaded hydrogels over the 10-day period (n = 3, technical replicates). (F) Schematic illustration showing that platelets cannot be effectively protected due to the chain rupture of the rigid hydrogels. (G) Compressive stress-strain curves of platelet-loaded F127DA hydrogel. (H) Dissipated energy and (I) loss tangent angle (tan δ) of the platelet-loaded hydrogels. (J–L) Finite element analysis of the platelet-loaded hydrogels. Data are presented as means ± S.E.M. (standard error of the mean). ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001, ns means no significance.

Moreover, NGF was rapidly released from the platelet-loaded PEGDA and GelMA groups after 600 cycles of compression, while the platelet-loaded F127DA group enabled sustained NGF release (Fig. 2C and Fig. S2). The release profiles of NGF from compressed PEGDA, GelMA, and F127DA hydrogels were fitted to five kinetic models (Table S1) [31]. The Hixson-Crowell model exhibited the highest correlation (R²), indicating that NGF release may be governed primarily by surface erosion and dissolution of the hydrogel matrix rather than diffusion [32,33]. These results further demonstrated the advantages of F127DA hydrogel in maintaining its structural integrity and controlling growth factor release under dynamic conditions.

Additionally, we assessed the protective effects of hydrogels on encapsulated platelets under dynamic culture conditions (PBS, 37 °C, 20 rpm orbital shaking). Our findings indicated that the majority of platelets in all hydrogels were activated over two weeks (Fig. S3). However, approximately 30 % of the platelets in the F127DA group still survived over 7 days, significantly higher than that in the other groups (p < 0.001) (Fig. 2D). The NGF release rates in this hydrogel over 20 days were obviously slower than those in other hydrogels (Fig. 2E). These results indicated that the F127DA hydrogel could protect platelets in dynamic culture conditions.

We further evaluated the protective efficacy of the hydrogel for platelets under in vivo conditions (Fig. S4). At 7 days post-subcutaneous implantation, the F127DA group exhibited a significantly higher platelet survival rate (22.2 ± 1.6 %) than both the PEGDA (7.4 ± 2.5 %, p < 0.001) and GelMA (9.3 ± 2.6 %, p < 0.001) groups, consistent with in vitro findings. Furthermore, SEM observation showed distinct platelet aggregation in the PEGDA and GelMA hydrogels, a key feature of activation, while the F127DA group retained clearly visible discoid-shaped, non-activated platelets (Fig. S5). These results indicated that the F127DA hydrogel could efficiently prevent platelets from being activated under both extreme and physiological situations. Moreover, we considered that the poor protective effect of PEGDA hydrogels on platelet activation might be attributed to its rigid internal networks, which were more prone to rupture under stress, leading to reduced energy dissipation (Fig. 2F) [34]. In contrast, GelMA hydrogel provides limited protection to platelets, thereby precluding its intrinsic protective effect on the platelets themselves.

To investigate the energy-dissipation behavior of the platelet-loaded F127DA hydrogel, we conducted load-unload tests at a strain range from 0 to 20 % (Fig. 2G and S6). The results showed that all platelet-loaded hydrogels exhibited hysteresis loops, while the platelet-loaded F127DA hydrogel displayed the most pronounced hysteresis behavior. Quantitatively, the dissipated energy of the platelet-loaded F127DA hydrogel exceeded 3.5 MJ/m³, outperforming the platelet-loaded PEGDA hydrogel (1.2 ± 0.2 MJ/m³) and GelMA hydrogel (0.7 ± 0.1 MJ/m³) (Fig. 2H). Then, the loss tangent angle (tan δ) was calculated from rheological test results (Fig. S7). The platelet-loaded F127DA hydrogel displayed a significantly higher tan δ than the other platelet-loaded hydrogels, demonstrating excellent damping capacity across a wide frequency range (Fig. 2I). These results confirmed that the platelet-loaded F127DA hydrogel exhibited outstanding energy-dissipation efficiency under compression.

Finite-element analysis was further performed to evaluate stress distribution within the hydrogels. The results revealed that the platelet-loaded F127DA hydrogel exhibited lower stress distribution under the same deformation compared to the PEGDA and GelMA hydrogels (Fig. 2J–L and S8). Commonly, hydrogels with reduced stress distribution exhibit enhanced energy dissipation capability, suggesting that these low-stress hydrogels are suitable for loading platelets [35]. F127DA molecules in the aqueous solution could self-assembled into 10–20 nm micelles through hydrophobic entanglement of the PPO segments (Fig. S9), and the hydrophobic core within these colloids could dissipate energy when exposed to the stress (Fig. 1) [27]. Our findings demonstrated that this nanocolloid hydrogel, with its superior energy-dissipation capacity, held great promise in ensuring the long-term survival of platelets.

The platelet-loaded energy-dissipative hydrogel was used to construct the functional nerve conduit. However, hydrogels often suffer from inadequate mechanical strength for surgical suturing. To develop a platelet-loaded conduit with sufficient mechanical strength, we established a novel method for rapidly bioprinting platelet-loaded hydrogel onto the surface of PCL nanofiber conduit (Fig. 3A and B). The effects of bioprinting parameters on platelets were evaluated by varying key processing conditions. The results showed a progressive reduction in platelet survival correlated with prolonged UV exposure, elevated rotational speed, and increased LAP concentration, likely resulting from enhanced shear stress and radical-induced activation (Fig. S10). Based on these findings, a parameter set of 30 s exposure, 60 rpm printing speed, and 0.75 % w/v LAP was selected for subsequent experiments, which could optimally balance between high platelet survival, printing efficiency, and structural fidelity of the conduit. SEM images showed that the PCL nanofibers in the inner layer of the PLT-F127DA conduit were radially aligned, while the morphology of the non-activated platelets in the outer layer appeared as discoid shape (Fig. 3C). The inner PCL nanofibers exhibited a porous structure with a pore size of 11.6 ± 1.4 μm and a high porosity of 80.2 ± 2.1 % (Table S2). The contact angle test also indicated that the dual-layered structure could improve the hydrophilicity of the inner layer in PLT-F127DA conduit. This improvement was attributed to the incorporation of hydrogel within the electrospun PCL network, which enhanced its hydrophilicity, a property conducive to cell adhesion (Fig. 3D and Fig. S11) [36].

Fig. 3.

Fabrication and characterization of platelet-loaded energy-dissipative hydrogel (PLT-F127DA) nerve conduits. (A) Schematic illustration of the conduit fabrication process. (B) Representative photographs of F127DA and PLT-F127DA conduits. (C) SEM images of PCL fibers and PLT-F127DA conduit: ⅰ) PCL fibers, ii) enlarged view of boxed region, iii) inner PCL fibers of conduit, iv) dual-layered conduit, v) platelets in outer hydrogel of conduit, vi) enlarged view of boxed region. (D) Contact angles of the PLT-F127DA conduits (n = 3, independent samples). (E) Photographs of the PLT-F127DA conduit after being folded and twisted. (F) Photographs of the tensile and compression tests of PLT-F127DA conduits. (G) Tensile strength and (H) Compressive strength of F127DA and PLT-F127DA conduits. (I–K) Finite element analysis of the PCL, F127DA, and PLT-F127DA conduits under a 0.5 N applied stress. (L) NGF and VEGF release profiles from the PLT-F127DA conduits at each time point (n = 3, technical replicates). (M) Schematic diagram of the dual-layered releasing model. (N) Release profiles of PLT (NGF)-loaded dual-layered releasing model (n = 3, technical replicates). (O) Swelling rate of different conduits (n = 3, independent samples). (P) Degradation rate of different conduits over 3 months (n = 3, independent samples). Data are presented as means ± S.E.M. (standard error of the mean).

This conduit also exhibited desirable mechanical properties, which could be easily twisted and folded (Fig. 3E). The suture retention test showed that the tensile strength of the PLT-F127DA conduit was 360.0 ± 5.4 Kpa (Fig. 3F and G), which close to the reported physiological range for native peripheral nerves (0.35–10 MPa in tensile strength) and was sufficient to meet surgical suturing requirements [[37], [38], [39], [40], [41]]. Meanwhile, the conduit was compressible with a stress strength of 110.3 ± 8.9 Kpa and could quickly recover its original shape after removing the stress (Fig. 3F and H). Additionally, finite element analysis was utilized to assess the compression-resistant performance of the PLT-F127DA conduit (Fig. 3I–K). When a force of 0.5 N was applied, the maximum deformation of the PLT-F127DA conduit (11.2 ± 0.5 %) was significantly lower than that of the F127DA conduit (54.5 ± 1.5 %) and PCL conduit (99.7 ± 0.2 %). Notably, the maximum deformation of the PLT-F127DA conduit was nearly an order of magnitude lower than that of the PCL conduit. These results demonstrated that the platelet-loaded dual-layered conduit had significantly enhanced mechanical strength.

The release of growth factors into the lumen of the conduit is crucial for nerve regeneration. We investigated the growth factor release from PLT-F127DA conduits. Sustained release of NGF and VEGF was observed (Fig. 3L), which could be released from both the inner layer and outer layer of the conduit. Given the critical role of lumen-directed release for nerve regeneration, we employed PLT and Cargo (NGF, VEGF) to prepare dual-layered releasing models (Fig. 3M). The results demonstrated sustained release of NGF and VEGF from the inner layer of the conduit over a period of one month (Fig. 3N and Fig. S12A), indicating that the majority of the growth factors could diffuse into the conduit lumen through the dual-layered structure. Notably, it was observed that the release of NGF and VEGF from the PLT-loaded releasing model was significantly more sustained than that from the model directly loaded with growth factors (Fig. S12B–C). Moreover, the concentrations maintained by the PLT-loaded conduit were found to align with previously reported levels required for effective nerve repair [42,43]. Furthermore, the release kinetics of NGF from the NGF-F127DA dual-layer conduit under simulated physiological conditions followed the Higuchi model over one month, as determined by the curve-fitting results presented in Table S3. This result further confirmed that the release mode of growth factors in conduits might mainly be the hydrogel-based diffusion [10]. The growth factors could sustainably diffuse out through the porous network of F127DA and PCL into the inner layer. Moreover, the PLT-F127DA conduit exhibited a relatively low swelling rate (<30 %), suggesting it can maintain a stable structure following in vivo implantation (Fig. 3O). The degradation rate of both conduits was nearly 20 % over 3 months (Fig. 3P). This degradation profile was sufficient to provide mechanical support during the critical early repair phase while allowing progressive remodeling with the host tissue [44].

2.2. The nerve conduits release growth factors to promote the cell proliferation and migration

The cytocompatibility of the nerve conduits was investigated by Live/dead and CCK8 assay. The results showed no significant differences in the survival and viability of SCs between the PLT-F127DA group and the Control group (Fig. 4A–C). Moreover, this conduit exhibited good hemocompatibility (Fig. 4D and E). We further accessed the effect of this conduit on the migration behaviors of HUVECs by scratch assays (Fig. 4F). After co-incubation for 72 h, the migration rate of the PLT-F127DA group (99.0 ± 1.0 %) was significantly higher than that of the Control group (63.7 ± 4.0 %, p < 0.001) (Fig. 4G), which might be attributed to the released VEGF from the slowly activated platelets. The results indicated that the conduits would promote angiogenesis for nerve regeneration. The effect of the conduits on SC migration was evaluated by Transwell migration assay. As shown in Fig. 4H, the PLT-F127DA conduit significantly promoted SC migration, which would facilitate the formation of Büngner bands after injury [45,46]. Furthermore, to better understand the interaction between PLT-F127DA and Schwann cells, we performed RT-qPCR (Fig. 4I–K). Compared to PCL and F127DA controls, PLT-F127DA conduits significantly upregulated mRNA expression of mTOR, c-Fos, Bcl-2, and Erk (p < 0.001, Fig. 4J and K), indicating TrkA pathway activation. Elevated c-Fos expression, an immediate-early gene, could directly confirm the transcriptional activation of downstream TrkA signaling [47]. These results suggested that the PLT-F127DA conduit enhanced Schwann cell-axon signaling to promote nerve regeneration and remyelination.

Fig. 4.

The biological activity of PLT-F127DA conduit in vitro. (A) Live/dead staining of SCs incubated with the leach liquor of F127DA and PLT-F127DA conduits and (B) quantification of cell survival rates (n = 3, technical replicates). (C) Viability of SCs after incubation with the leach liquor of conduits (n = 4, technical replicates). Representative images (D) and quantitative analysis (E) of the hemolysis test for F127DA and PLT-F127DA conduits (n = 4, technical replicates). (F) Migration of HUVECs after incubation with the leach liquor of conduits. (G) Statistical analysis of the HUVEC migration rates (n = 3, technical replicates). (H) Migration of SCs after incubation with the leach liquor of conduits. (I) Schematic illustration of the dual-layered PLT-F127DA model co-cultured with SCs in a Transwell plate. (J) RT-qPCR analysis of SCs co-cultured with various conduits (n = 3, technical replicates). (K) Schematic illustration of TrkA pathway activation by the PLT-F127DA. Data are presented as means ± S.E.M. (standard error of the mean). ∗p < 0.05, ∗∗∗p < 0.001, and ns means no significance.

To further confirm the benefits of long-acting platelets in cellular experiments, we observed 14-day long-term neurite outgrowth of PC12 cells to distinguish the independent contributions of typical cargo (NGF and VEGF), free platelets, the hydrogel matrix, and their synergistic effects [48]. The experimental groups were set as follows: control group (Control), cargo group (NGF and VEGF), free platelets group (PLT), platelet-free F127DA hydrogel group (F127DA), and platelet-loaded F127DA hydrogel group (PLT-F127DA). PC12 cells in the PLT-F127DA group exhibited a more extended and oriented morphology within 2 weeks (Fig. 5A and B). On day 7, statistical analysis revealed that neurite length in the PLT-F127DA group (76.0 ± 4.3 μm) was significantly greater than that in the F127DA group (45.6 ± 3.1 μm, p < 0.001), but showed no significant difference compared to the Cargo group (78.8 ± 1.5 μm, p > 0.05) and PLT group (82.4 ± 2.6 μm, p > 0.05) (Fig. 5C). By day 14, neurite length in the PLT-F127DA group (108.2 ± 3.0 μm) significantly exceeded that of the F127DA group (68.3 ± 3.2 μm, p < 0.001), Cargo group (85.9 ± 4.6 μm, p < 0.001), and PLT group (97.1 ± 4.5 μm, p < 0.05) (Fig. 5D). The results demonstrated that although free platelets or the direct addition of an amount of growth factors equivalent to that released by platelets provide an initial stimulatory effect, their benefits were transient due to rapid activation and subsequent degradation of the growth factors, which failed to sustain the prolonged signaling required for robust neural regeneration [20]. In contrast, the PLT-F127DA group showed a stronger and more sustained promotion of neurite outgrowth and orientation. Moreover, the proportion of oriented PC12 cells in the PLT-F127DA group exceeded 80 %, while those of the other groups were below 50 % (Fig. 5E–I). These results suggested that the PLT-F127DA nerve conduit could facilitate axon elongation and directional growth through its sustained release of abundant growth factors (Fig. 5J).

Fig. 5.

Bioactive effects of PLT-F127DA conduits on PC12 cells in vitro. (A–B) Actin cytoskeleton staining of PC12 cells after incubation with conduits for 7 and 14 days. Quantification of PC12 cells neurite length at day 7 (C) and day 14 (D) (n = 3, technical replicates). (E–I) Polar histograms of neurite orientation in PC12 cells cultured across conduits at day 14 (n = 3, technical replicates). (J) Schematic illustration of PC12 cells cultured on dual-layered PLT-F127DA model. Data are presented as means ± S.E.M. (standard error of the mean). ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001, ns means no significance.

To investigate the performance of platelet-loaded conduits on cell proliferation and migration at the early stage, a 10 mm sciatic nerve defect model of rat was established. The immunostainings revealed that a large number of cells migrated and proliferated into the conduit lumen from the injured stumps in all groups (Fig. 6A). Notably, the distances between the front of the regenerated cells and the injured stump were significantly greater in the PLT-F127DA group compared to the other groups. The positive areas of NF200 and S100 in the PLT-F127DA group were more substantial at 1-week post-operation, compared to the PLT-PEGDA group and PLT-GelMA group. These results suggested that the PLT-F127DA conduit could promote cell infiltration, proliferation, and migration at the early stage of regeneration.

Fig. 6.

Evaluation of nerve conduits in promoting early nerve regeneration. (A) Immunostainings of NF200 (green), S100-β (red), and DAPI (blue) in longitudinal sections of platelet-loaded PEGDA, GelMA, and F127DA conduits at 1 week post-surgery. (B) H&E staining images of regenerated nerve longitudinal sections at 3 weeks post-surgery. The immunofluorescence images of (C) NF-200 (red), S100-β (green), DAPI (blue), and (D) CD31 (red) at 12 weeks post-surgery. Percentage of (E) NF-200 and (F) S100-β positive areas in regenerated nerves (n = 3, technical replicates). (G) Blood vessel density in regenerated nerves (n = 3, technical replicates). Data are presented as means ± S.E.M. (standard error of the mean). ∗∗p < 0.01, ns means no significance.

Subsequently, we evaluated early nerve regeneration of the PLT-F127DA, F127DA, and Auto groups at 3 weeks post-operation. As shown in Fig. 6B, more regenerated tissues extended from the proximal end to the distal end along the channel in the PLT-F127DA conduit. The immunofluorescence images and quantitative analysis (Fig. 6C and 6E-F) showed that the percentage of NF200 and S100β positive areas in the PLT-F127DA group were notably higher than those of the F127DA group (p < 0.01), while close to that of the Auto group (p > 0.05). Additionally, immunostaining of CD31 revealed that the blood vessel density in the PLT-F127DA group (39.00 ± 3.61/mm²) was significantly higher than that in the F127DA group (20.00 ± 4.36/mm², p < 0.01), and comparable to the Auto group (35.00 ± 2.00/mm², p > 0.05) (Fig. 6D and G). These findings indicated that PLT-F127DA conduit might provide a growth factor-rich microenvironment that facilitate early-stage nerve repair by promoting axonal regeneration, remyelination, and angiogenesis.

2.3. Nerve repair using the platelets-loaded nerve conduits in vivo

We further investigated the therapeutic efficacy of the platelet-loaded conduits (Fig. 7A). At 12 weeks post-implantation, the implanted conduits were partially degraded but still maintained their structural integrity (Fig. 7B and Fig. S13). The functional recovery of the regenerated nerves was evaluated by gait analysis (Fig. 7C). The results showed that the SFI value of the PLT-F127DA group was −58.30 ± 5.23, which was significantly higher than that of the F127DA group (−78.50 ± 2.35, p < 0.001) and had no significant difference with the Auto group (−52.05 ± 5.61, p > 0.05) (Fig. 7D). Additionally, electrophysiological analysis at the injured site was further examined. The NCV value of the PLT-F127DA group was 48.76 ± 7.71 m s⁻¹, comparable to that of the Auto group (49.58 ± 5.05 m s⁻¹, p > 0.05) (Fig. 7E and F). The latency CMAP onset in the PLT-F127DA group (1.57 ± 0.36 ms) was close to that of the Auto group (1.50 ± 0.40 ms, p > 0.05) (Fig. 7G). Moreover, the reinnervation of the gastrocnemius muscle was assessed (Fig. 7H–K). The muscle was evidently atrophied in the PLT-F127DA, F127DA, and Auto groups compared to the Sham group (Fig. 7H). The relative muscle wet weight of the PLT-F127DA group was significantly higher than that of the F127DA group (p < 0.05), while comparable to that of the Auto group (p > 0.05) (Fig. 7I). Meanwhile, as shown in Fig. 7J and K, the PLT-F127DA group (129.0 ± 18.2 μm) and the Auto group (126.5 ± 19.2 μm) both exhibited a significantly larger average muscle fiber diameter compared to the F127DA group (87.6 ± 23.6 μm, p <0.001). No significant difference was observed between the PLT-F127DA and the Auto groups (p > 0.05). These results revealed that the PLT-F127DA nerve conduit positively improved the functional recovery of injured nerves.

Fig. 7.

Functional restoration of injured nerves at 12 weeks post operation. (A) Schematic illustration showing the process of nerve repair. (B) Representative images of regenerated nerves. (C) Walking footprints of SD rats, the distance between digits of denervated paws was reduced. (D) SFI values of the sciatic nerves calculated from walking footprints (n = 5, biologically independent animals per group). The SFI ranges from −100 to 0: −100 indicates complete paralysis of the hind limbs, and 0 indicates normal. (E) Representative CMAP recordings. (F) NCV and (G) latency of CMAP onset at the defect site (n = 3, biologically independent animals per group). (H) The photographs of the gastrocnemius muscles. (I) The quantitative analysis of relative wet weight of gastrocnemius muscle (n = 3, biologically independent animals per group). (J) H&E images of the sections of gastrocnemius muscle. (K) Quantitative assessment of mean muscle fiber diameter (n = 60, technical replicates). Data are presented as means ± S.E.M. (standard error of the mean). ∗p < 0.05, ∗∗∗p < 0.001, ns means no significance.

Moreover, immunostainings and histological assessments were performed to evaluate the morphology of the distal segments of the regenerated nerves. The immunostainings and quantitative analysis (Fig. 8A and B) showed that the percentage of GAP43 positive areas in the PLT-F127DA group was significantly higher than that of the F127DA group (p < 0.01), while close to that of the Auto group (p > 0.05). Additionally, immunostaining of CD31 revealed that the blood vessel density in the PLT-F127DA group (47.67 ± 4.16/mm²) was notably higher than that in the F127DA group (34.67 ± 2.52/mm², p < 0.01), and comparable to the Auto group (45.00 ± 2.00/mm², p > 0.05) (Fig. 8C and D). H&E staining revealed that axons originating from the proximal end successfully regenerated to the distal end in all groups (Fig. S14). Numerous well-aligned myelinated axons were observed by LFB staining (Fig. 8E). Quantitatively, the myelinated axon density in the PLT-F127DA group was impressively higher than that of the F127DA conduit group (p < 0.01), and comparable to that of the Auto group (p > 0.05) (Fig. 8F). Additionally, the ultrastructure of the myelinated axons was observed by TEM (Fig. 8G). The myelin sheaths in each group were dispersed into clusters, displaying a uniform structural organization. The myelin sheath thicknesses of the PLT-F127DA, F127DA, and Auto groups were 0.57 ± 0.05 μm, 0.43 ± 0.04 μm, and 0.57 ± 0.04 μm, respectively (Fig. 8H). The numbers of myelin layers were 34.67 ± 3.05, 27.67 ± 4.16, and 35.67 ± 3.79, respectively (Fig. 8I), and the axon diameters were 3.35 ± 0.37 μm, 2.72 ± 0.44 μm, and 3.67 ± 0.45 μm, respectively (Fig. 8J). There were no significant differences in myelin thickness, number of myelin layers, or axon diameter between the PLT-F127DA group and Auto group (p > 0.05), and both were higher than those of the F127DA group (p < 0.05). Lastly, histological analysis further confirmed that both F127DA and PLT-F127DA conduits exhibited no adverse effects on the major organs of the rats (Fig. S15). These findings indicated that the PLT-F127DA conduit could effectively protect the encapsulated platelets from activation, thereby facilitating axon regeneration and remyelination.

Fig. 8.

Morphological evaluation of regenerated nerves after twelve weeks. (A) Immunofluorescence images of GAP43 (green) and DAPI (blue) in cross-sections of Sham, Auto, F127DA and PLT-F127DA groups. (B) Percentage of GAP43 positive areas in regenerated nerves (n = 3, technical replicates). (C) Immunofluorescence images of CD31 (red) and DAPI (blue) in cross-sections of all groups. (D) Blood vessel density in regenerated nerves (n = 3, technical replicates). (E) Photographs of myelinated nerves and (F) density of myelinated nerve fibers (n = 3, technical replicates). (G) TEM images of regenerated nerves. Quantitative analysis of (H) myelin sheath thickness (n = 3, technical replicates), (I) numbers of myelinated sheath layers (n = 3, technical replicates), and (J) axons diameter (n = 6, technical replicates). Data are presented as means ± S.E.M. (standard error of the mean). ∗p < 0.05, ∗∗p < 0.01, ns means no significance.

3. Discussion

Platelets show significant potential in tissue repair by releasing growth factors in situ. The sustained release of these growth factors is beneficial for treatment. Hydrogels have potential in protecting platelets from fast activation and cargo release [17]. Commonly, these hydrogels are soft and perform well in static environments [49]. However, platelet-loaded hydrogels often suffer from deformation, which can induce stress and subsequently activate the platelets. In recent years, hydrogels systems such as PEGDA, GelMA, and alginate have been used to encapsulate platelets or platelet derivatives and reduce premature activation [49,50]. However, these materials often suffer from fragile network structures or insufficient mechanical stability, making them prone to deformation in dynamic physiological environments and thereby risking damage to the encapsulated platelets. To address this issue, we focused on the effects of local energy distribution on platelet activation under external stress [24,25]. Herein, we used an energy-dissipative hydrogel to improve the survival of platelets during deformation. Our results indicated that this hydrogel with nanostructured energy-dissipation centers could significantly protect platelets under both extreme and dynamic conditions. This work might be the first attempt to use the energy-dissipative hydrogel to protect platelets from rapid activation and to develop a long-acting platelet preparation.

While platelet-loaded hydrogels hold great potential for long-term release of growth factors in nerve repair, their application in nerve conduits has often been restricted by inadequate mechanical strength [50,51]. Multiple strategies have been developed to improve the mechanical properties of hydrogels, but most of them suffer from the risk of rapid platelet activation [23,52]. In this study, we developed a dual-layered nerve conduit, in which the outer hydrogel layer could effectively protect platelets, while the inner electrospun PCL nanofiber scaffold provided resistance to stretching. Different from the reported dual-layered conduit structure using hydrogel as the inner layer [53], our dual-layered structural design could allow the bioprinting of platelet-loaded hydrogels onto electrospun fibers, minimizing the risk of platelet activation. However, fabricating conduits with this structural configuration posed challenges for existing bioprinting technologies, including slow printing speed, material compatibility issues, and insufficient structural precision [54]. To overcome these limitations, we developed a customized DLP-based rapid printing system that could fabricate the conduit within 30 s, which not only greatly shortened the fabrication time but also maximized the preservation of platelet activity. This strategy provides a new design concept for the application of platelet-hydrogel systems in nerve conduits and lays the foundation for the future development of personalized hydrogel conduits with both bifunctionality and mechanical stability.

Previous studies have indicated that conduits delivering multiple growth factors are more effective in promoting nerve regeneration and functional recovery than those delivering a single growth factor [55]. In this study, the PLT-F127DA conduit could deliver multiple growth factors with a formulation resembling physiological conditions. Moreover, the PLT-F127DA conduit exhibited a superior capacity for prolonging the survival of platelets, enabling the sustained release of multiple growth factors to provide biological cues for guiding nerve regeneration. Previous research has also shown that the quality and speed of early nerve regeneration are crucial for the morphological and functional recovery of injured nerves [56,57]. Our results demonstrated that this PLT-F127DA conduit, with excellent mechanical strength and biodegradability, could facilitate axonal regeneration and functional recovery comparable to autografts. This conduit with prolonged growth factor release has potential clinical application in the treatment of peripheral nerve defects. Despite these promising regenerative outcomes, the clinical translation of the PLT-F127DA conduit still faces challenges such as limitations in scalable manufacturing, insufficient long-term safety evidence, and lack of standardized platelet sources. Future work should focus on large-animal validation, enhancing material biocompatibility and degradation control, and integrating imaging-based personalized design to facilitate clinical application [[58], [59], [60]].

4. Conclusions

In this work, we developed a platelet-loaded nerve conduit for the effective peripheral nerve repair. We discovered that the energy-dissipative hydrogel could protect platelets from rapid activation, enabling the prolonged release of cargo even under deformation-induced stress. It offers a reliable method for constructing a biosynthetic nerve conduit by coating platelet-loaded soft hydrogels onto an electrospun conduit via a fast bioprinting process. The obtained nerve conduit with excellent mechanical properties could efficiently repair nerve defects by providing a growth factor-rich microenvironment, showing potential for clinical applications. While the in vivo results are promising, further validation in larger animal models remains essential. Subsequent studies will focus on the long-term behavior of the platelets and the scalability of the manufacturing process.

5. Materials and methods

5.1. Materials

Materials included β-NGF (Novoprotein, China), live/dead cell kit (Thermo Fisher Scientific, USA), Cell Counting Kit-8 (CCK-8) (MCE, USA), Actin-Tracker Green-488 (Beyotime, China), 4 % paraformaldehyde (PFA) (Biosharp, China), Rat ELISA Kit (Ruixin Biotechnology Co., Ltd., China), Animal Total RNA Isolation Kit (Foregene, China), cDNA Synthesis SuperMix (Yeasen, China), Universal Blue qPCR SYBR Green Master Mix (Yeasen, China) and Polyethylene glycol diacrylate (PEGDA) (Macklin, China). Other chemical reagents were sourced from Sigma Aldrich (USA), except for Pluronic F127 diacrylate (F127DA), Gelatin methacryloyl (GelMA), and lithium phenyl-2,4,6-trimethyl-benzoylphosphinate (LAP), which were synthesized following previously published methods [30].

5.2. Cells and animals

Schwann cells (SCs), Human Umbilical Vein Endothelial Cells (HUVECs), and PC12 cells were obtained from BeNa Culture Collection (China). Cells were maintained in Dulbecco's Modified Eagle Medium (DMEM) (Sigma Aldrich, USA), enriched with fetal bovine serum (FBS) (Thermo Fisher Scientific, USA) at a final concentration of 10 %, along with Penicillin (100 U mL ⁻¹) and Streptomycin (100 mg mL ⁻¹).

Male SD rats (5 weeks old) and GFP transgenic C57BL/6-Tg (CAG-EGFP) mice were provided by HFK Bio-Technology Co. Ltd. (China) and were kept in an environment free of specific pathogens. All animal studies were conducted under protocols approved by the Animal Research and Ethics Committee of Sichuan University (approval number: 20241112008).

5.3. Platelets isolation

Blood samples were obtained from 8-week-old male SD rats through abdominal aorta puncture or from GFP transgenic C57BL/6-Tg mice via surgical eye enucleation. For platelets isolation, the collected blood underwent an initial centrifugation at 1000 rpm for 15 min. Subsequently, the supernatant was harvested and subjected to a second centrifugation at 100×g for 8 min. Lastly, the residual supernatant was centrifuged at 2000 rpm for 15 min to effectively precipitate the platelets.

5.4. Preparation and characterization of hydrogels

Cylindrical hydrogels (Φ25 mm × 1 mm) were prepared by photopolymerization as samples for dynamic rheological and compressive tests with an MCR 302e rheometer (Anton Paar, Austria). The hydrogels were prepared by mixing the prepolymer solution (PEGDA, GelMA, F127DA, 15 % w/v) with or without platelets (1 × 10⁸ platelets mL⁻¹). Amplitude sweep (10 rad s⁻¹) test was performed to determine the linear viscoelastic region. Among them, a constant strain (0.1 % strain) within the linear range was selected for frequency sweeps to measure the storage modulus (G′) and loss modulus (G″) of the samples. The formula for calculating the dissipated energy of different platelet-loaded hydrogels is as follows [61]:

$$E_{\text{dissipated}}=\underset{\text{loading}}{\int }\sigma d\epsilon -\underset{\text{unloading}}{\int }\sigma d\epsilon$$

Where E represents the dissipated energy (MJ/m³), σ is the stress (MPa), and ε is the strain (%). The area under the loading curve minus the area under the unloading curve equals the energy dissipation. The loss tangent angle (tan δ) is represented as follows:

$$\mathrm{Tan}\delta ={\mathrm{G}}^{″}/{\mathrm{G}}^{\prime }\times 100\%$$

Finite element simulations were conducted using ANSYS software. The mechanical properties of the hydrogel were defined based on the stress-strain curves obtained from compressive tests. The hydrogel sample geometry was modeled in ANSYS Workbench. Hexahedral meshing with an element size of 0.3 mm was implemented to ensure computational reliability, followed by application of 20 % compressive strain in the vertical direction to simulate uniaxial compression behavior. To observe the microstructure of F127DA solution, the sample was deposited on a carbon-coated copper grid and observed using a Transmission Electron Microscope (TEM; Hitachi, Japan) at a voltage of 80 kV.

5.5. Survival rates of platelets in hydrogels

To assess the protective effect of hydrogels on platelets in vitro, the extracted GFP-labeled platelets were mixed with prepolymer solution (PEGDA, GelMA, F127DA, 15 % w/v) and crosslinked to cylindrical hydrogels under UV light exposure. Platelet cyclic assessments were conducted using a load of 20 g (Fig. S1). Meanwhile, the survival of platelets in the hydrogels was determined by a Laser Scanning Confocal Microscope (LSCM) (Nikon Corporation, Japan). Additionally, hydrogels were incubated in a plate with 100 μL PBS solution at 37 °C under shaking at 20 rpm to evaluate their protective efficacy.

To evaluate the protective effect of hydrogels on platelets in vivo, GFP-labeled platelet-loaded hydrogels were implanted subcutaneously into the dorsal side of mice (n = 3 per group). On day 7, the implantation sites were exposed, and the hydrogels were harvested and observed by LSCM at predetermined time points. Images were acquired, and platelet survival rates were quantitatively analyzed using ImageJ software. Meanwhile, the platelet-loaded hydrogels were cryo-fractured in liquid nitrogen, dehydrated through a graded ethanol series, and critical-point dried. Subsequently, the samples were sputter-coated with gold and observed using a scanning electron microscope (SEM; Thermo Fisher Scientific, USA).

During the fabrication of nerve conduits using rapid printing technology, GFP-labeled platelets were isolated and incorporated into the prepolymer solutions (PEGDA, GelMA, F127DA, 15 % w/v) to systematically evaluate the influence of printing parameters on platelet survival and function. Key printing parameters, including UV exposure time (10, 20, and 60 s), printing rotation speed (20, 60, and 120 rpm), and photoinitiator concentration (LAP at 0.5 %, 0.75 %, and 1 % w/v), were systematically examined. Ultimately, we identified conditions that optimally balanced platelet survival, printing efficiency, and structural fidelity of the conduits.

5.6. Fabrication and characterization of nerve conduits

Electrospinning solutions were prepared by dissolving 6 % (w/v) polycaprolactone (PCL) in 1,1,1,3,3,3-hexafluoro-2-propanol. The electrospinning parameters involved a positive voltage set at 8 kV, a negative voltage at 5 kV, and a gap of 150 mm maintained between the needle tip and the cylindrical receiver (Φ1.8 mm). Then, the receiver with the PCL conduit was fixed to a customized sample tank (Φ2.3 mm) using a self-developed clamping system connected to a computer-controlled rotation device integrated with the printer software. Both ends of the receiver were firmly held by the clamps to ensure stability during the process. Besides, 200 μL platelet-loaded hydrogel was added evenly to the surface of PCL conduit. Next, the 3D digital model of the conduit, designed using SolidWorks, was imported into slicing software to generate the corresponding layer-by-layer sliced images. These images were then input into a light engine (LightCrafter 4500) to form a dynamic digital light beam, which was projected onto the surface of the PCL conduit. The printing parameters were set as follows: printing time of 30 s, rotation speed of 60 rpm. Furthermore, the printed nerve conduits were soaked in PBS to remove the LAP and un-crosslinked polymers.

The contact angles of the samples were tested by contact angle measurement instrument (JY-Pha, China). The tensile and compressive properties of the samples were measured by a Dynamic Mechanical Analyzer (DMA) (TA Instruments, USA), with a force of 0.5 N/min. Meanwhile, the PCL fibers, the frontal and cross-sectional micromorphology of the platelet-loaded energy-dissipative hydrogel conduits were coated with gold and observed using a scanning electron microscope (SEM) (Thermo Fisher Scientific, USA). The pore size of the PCL conduit was determined by analyzing the SEM images with ImageJ software, while the overall porosity was measured via the liquid displacement method. The samples were immersed in PBS for 24 h, and the swelling rates were assessed as the ratio of the finial weight to the original weight. Furthermore, finite element simulations were conducted using ANSYS software. The conduit geometry was modeled in Workbench and discretized with hexahedral elements at a mesh size of 0.3 mm. Subsequently, a 0.5 N compressive load was applied vertically to simulate uniaxial compression behavior. The gravimetric analysis was conducted to evaluate the in vitro degradation rate of nerve conduits. The conduits were weighed and then immersed in PBS solution at 37 °C under shaking at 100 rpm. At predetermined intervals, these samples were rinsed with deionized water, dried and accurately weighed, and the percentage of residues can be calculated from the residual weight and initial weight.

5.7. Release profiles of growth factors

Both compressed and uncompressed platelet-loaded hydrogels were incubated in PBS at 37 °C, with unstressed samples additionally subjected to 20 rpm orbital shaking. At specified intervals, the releasing solution was harvested and preserved at −20 °C. The amount of released NGF was then tested by Rat Nerve Growth Factor (NGF) ELISA Kit. Meanwhile, the release kinetics of the platelet-loaded hydrogels following compression were further analyzed by fitting the data to five mathematical models: zero-order, first-order, Higuchi, Hixson–Crowell, and Korsmeyer–Peppas [31]. The growth factor release solution of PLT-F127DA conduits was obtained by the same methods, and the amount of released NGF and VEGF were tested with Rat Nerve Growth Factor ELISA Kit.

The platelets and cargo (NGF and VEGF) were used to conduct experiments on the direction of factor release. The PCL films were trimmed into squares and placed close to the filter membrane of the upper chamber of the 24-well Transwell plate. The platelet-loaded F127DA solution, or a platelet-free F127DA solution containing equivalent doses of NGF and VEGF, was then added inside the PCL film and immediately photopolymerized. Fresh simulated body fluid (SBF) was added to the lower chamber of the plate, which was then placed in a shaking incubator maintained at 37 °C. At specified intervals, the releasing solution was collected for analysis, and an equivalent volume of fresh SBF was replenished. Meanwhile, the amounts of released NGF and VEGF were tested with Rat Nerve Growth Factor ELISA Kit. Subsequently, the release kinetics of the NGF-loaded F127DA releasing model were analyzed by fitting the experimental data to five mathematical models: zero-order, first-order, Higuchi, Hixson–Crowell, and Korsmeyer–Peppas [31].

5.8. Biocompatibility assessments of nerve conduits

The leach liquor of the conduits was prepared by incubating them with DMEM medium for 24 h at 37 °C. SCs were seeded into a 96-well plate (10⁴ cells per well) for 24 h. Then, the supernatant was removed, and the leach liquor was added into plates. Subsequently, the cells were incubated for 24, 48 and 72 h. The cell vitality was evaluated by the CCK-8 assay. Besides, the cells were incubated with Calcein AM/PI and observed by Live-Cell Imaging Microplate Detection System (Olympus Corporation, Japan) to evaluate the survival percentage. To further confirm the in vitro biocompatibility of the conduits, the hemolysis assessment was performed. Fresh blood, ultrapure water, normal saline, and normal saline extracts with the treatment of PLT-F127DA and F127DA groups (5 mg mL⁻¹) were pipetted 2 mL into the 5 mL centrifuge tubes, respectively. The fresh blood was diluted with 1.25 times normal saline. Subsequently, prewarmed diluted blood was added to each group and incubated at 37 °C for 1 h, followed by centrifugation at 3000 rpm for 5 min. Finally, the optical density (OD) of the supernatant was measured spectrophotometrically at 545 nm, and the hemolysis rate for the samples was calculated from the equation below:

$$\text{Hemolysis}\text{rate}(\%)={\text{OD}}_{\mathrm{n}}–{\text{OD}}_1/{\text{OD}}_2-{\text{OD}}_1\times 100\%$$

where OD_n, OD₁, OD₂ represent the OD values of the normal saline extract with the conduits, the blank normal saline, the ultrapure water, respectively.

5.9. Promotion of cell migration by nerve conduits

The platelet-free 127DA conduit and the platelet-loaded PLT-F127DA conduit were prepared as described in Section 5.6 for subsequent functional evaluation via a cell migration assay. To obtain leach liquors, both the F127DA and PLT-F127DA conduits were individually immersed in DMEM medium at a ratio of 0.5 cm² mL⁻¹ for 24 h at 37 °C. Fresh DMEM medium and the leach liquors from the platelet-free F127DA conduit were used as control groups to assess the specific effects of growth factors released from the PLT-F127DA conduit on cell behavior. HUVECs were seeded into 6-well plate (2 × 10⁵ cells per well) and incubated for 12 h. Then, the cells were scratched by a pipette tip. The leach liquors were introduced to the cells, and the samples were incubated for predetermined time points. The cells were observed via optical microscopy at 0, 24, 48, and 72 h, respectively. The scratch wound areas were quantitatively analyzed using ImageJ software.

For the migration of SCs, cells were plated into the 24-well Ttranswell plate upper chamber (5 × 10⁴ cells per well) and co-cultured with FBS-free DMEM medium. Besides, the previous prepared leach liquors were added to the lower chamber of the Ttranswell plate, and the DMEM was the control group. Subsequently, after fixing and washing, the cells were stained by 0.1 % crystal violet at room temperature for 20 min. Finally, the migration of cells was imaged using inverted optical microscopy.

5.10. RT-qPCR analysis

The platelet-free 127DA conduit and the platelet-loaded PLT-F127DA conduit were prepared following the protocol outlined in Section 5.6 for RT-qPCR analysis. Schwann cells (2 × 10⁵ cells/well) were co-cultured with nerve conduits using a 6-well Transwell plate (Corning, 0.4 μm pore). After 72 h, total RNA was extracted for RT-qPCR analysis with an Animal Total RNA Isolation Kit. Then, cDNA was synthesized using a cDNA Synthesis SuperMix, followed by qPCR with Universal Blue SYBR Green Master Mix on a QuantStudio Real-Time PCR System (Thermo Fisher Scientific, USA). All the primers were listed in Table 1, and GAPDH was used as reference gene.

Table 1.

Real-time PCR primers.

Gene	Primers Froward (3′→5′)	Primers Reverse (5′→3′)
Bcl-2	CTTCTCTCGTCGCTACCGTC	CAATCCTCCCCCAGTTCACC
mTOR	GCAATGGGCACGAGTTTGTT	AGTGTGTTCACCAGGCCAAA
c-Fos	GGGAGCTGACAGATACGCTC	ATTGGCAATCTCGGTCTGCA
Erk	GGCATCCGAGACATCCTCAG	TATGTACTTGAGGCCCCGGA
GAPDH	GACATGCCGCCTGGAGAAAC	AGCCCAGGATGCCCTTTAGT

5.11. Neurite outgrowth in PC12 cells

The platelet-free 127DA conduit and the platelet-loaded PLT-F127DA conduit were prepared following the protocol outlined in Section 5.6 for subsequent functional evaluation using a neurite outgrowth assay. Fresh DMEM medium, Cargo (NGF and VEGF), free platelet (PLT), and platelet-free 127DA conduit were used as control groups to assess the specific effects of growth factors released from the PLT-F127DA conduit on cell behavior. PC12 cells were co-incubated with the conduits at a density of 1 × 10³ cells in 35 mm confocal dishes. On day 7 and day 14, the cells were fixed in 4 % PFA for 1 h and subsequently washed with PBST solution (0.1 % Triton X-100 in PBS). The samples were then incubated with a primary antibody against TUJ1 (1:250, Proteintech) overnight at 4 °C, followed by incubation with a Cy3-conjugated goat anti-mouse IgG (H + L) secondary antibody (1:50, Proteintech) for 1 h at 37 °C. Subsequently, actin filaments were stained using Actin-tracker green-488 (1:200, Beyotime) for 1 h, following washing with PBST. Finally, the samples were stained by DAPI for 10 min and then imaged by a LSCM.

5.12. Surgical process

180–200 g healthy male SD rats were divided into pre-defined groups to evaluate the therapeutic efficiency of the conduits. To investigate nerve regeneration at early stage (1 week) within different platelet-loaded hydrogel conduits, rats were randomly assigned to three groups (n = 4 per group): PLT-PEGDA conduit, PLT-GelMA conduit, PLT-F127DA conduit. To evaluate the therapeutic efficacy of the conduits at 3- and 12-week, rats were randomly divided into four groups (n = 10 per group): sham-operation (Sham), autograft (Auto), F127DA conduit, and PLT-F127DA conduit. Briefly, under pentobarbital sodium anesthesia (45 mg/kg, i.p.), a 10-mm sciatic nerve defect was surgically created in the right hindlimb of each rat, and the defect sites were implanted with the corresponding grafts.

5.13. In vivo morphological evaluation of regenerated nerves

Regenerated sciatic nerves were harvested at pre-defined time points post-surgery and fixed in 4 % PFA. After embedding in OCT (SAKURA, USA), the samples were sectioned longitudinally or transversely according to the experimental requirements, and then stained with H&E. For the longitudinal sections obtained at the first and third weeks, immunostaining was performed. The sections were incubated with mouse anti-NF-200 antibody (1:400, Sigma Aldrich) and rabbit anti-S100-β antibody (1:100, Abcam) overnight at 4 °C, followed by incubation with goat anti-mouse IgG-Alex-488 (1:1000, Abcam) and goat anti-rabbit IgG-Alex-594 (1:1000, Abcam) for 1 h at room temperature. The nuclei were stained by DAPI for 5 min. Additionally, sections were incubated with rabbit anti-CD31 antibody (1:1000, Proteintech) using the same protocol to assess angiogenesis. Finally, the samples were observed via Fully Automated Pathology Imaging and Analysis System VS200 (Olympus, Japan).

After 12 weeks post-surgery, distal segments of regenerated nerves were harvested, cross-sectioned, and processed for immunostaining against GAP-43 (1:100; Proteintech) and CD31. TEM (HITACHI, Japan) and histological analyses (H&E and modified Loyez staining) were subsequently performed to evaluate myelinated axon regeneration at the repair site. Meanwhile, the major organs harvested from the rats were also processed for H&E staining.

5.14. Statistical analysis

Statistical evaluations were performed using either Student's t-test or ANOVA, with subsequent Bonferroni post hoc analysis (GraphPad Prism). Results were presented as mean ± standard error of the mean (S.E.M.). Significance levels were indicated as ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

5.15. In vivo functional restoration and degradation test

To assess the recovery of motor function, the gait parameters of the rats at 12 weeks after operation were assessed via sciatic nerve functional index (SFI). The SFI values were calculated using the following formula [62]:

$$\text{SFI}=109.5\left(\text{ETS}–\text{NTS}\right)/\text{NTS}–38.3\left(\text{EPL}–\text{NPL}\right)/\text{NPL}+13.3\left(\text{EIT}–\text{NIT}\right)/\text{NIT}–8.8$$

Where TS is the toe spread, PL is the print length, and E and N refer to the experimental and normal contralateral hind paws, respectively.

After 12 weeks of surgery, an electrophysiological test was processed to evaluate the efficacy of the conduits in bridging the sciatic nerve gaps. The gastrocnemius muscles were photographed and weighed to calculate their relative wet weight. Then, the muscles were cross-sectioned for H&E staining, and the diameter of muscle fibers was determined by ImageJ software. Moreover, the degradation of the conduits (13 mm) was assessed by subcutaneously implanting them on the backs of SD rats. After 1,2, and 3 months of surgery, the implanted conduits and adjacent tissues were collected for H&E staining.

CRediT authorship contribution statement

Jiamei Zhang: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Conceptualization. Wenbi Wu: Writing – original draft, Investigation, Formal analysis. Ya Ren: Writing – review & editing, Investigation, Formal analysis. Xide Dai: Methodology, Investigation. Shuwei Ye: Methodology, Investigation. Wei Zhao: Methodology. Haofan Liu: Methodology. Liming He: Methodology. Boya Li: Writing – review & editing. Li Zhang: Resources. Xia Luo: Resources. Wentao Li: Resources. Xue Zhang: Resources. Shuai Yang: Resources. Maling Gou: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

Ethics approval and consent to participate

This study and included experimental procedures were approved by the Animal Research and Ethics Committee of Sichuan University (approval No. 20241112008). All animal housing and experiments were conducted in strict accordance with the institutional guidelines for care and use of laboratory animals.

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 is the Supplementary data to this article:

Data availability

Data will be made available on request.