Engineered decellularized muscle extracellular matrix fabrics enable functional volumetric muscle loss repair Authors: Guangzhou Song, Wenqian Cong, Hongbai Lu, Jiatong Wang, Yumeng Wang, Yanzhen Zhao, Shaowen Wang, Shijie Zhu, Adam C Midgley, Wen Li, Meifeng Zhu Published: 2026 ABSTRACT AbstractVolumetric muscle loss (VML) presents a significant clinical challenge, as existing biomaterials fail to recapitulate the anisotropic architecture and pro-regenerative niche that is essential for functional muscle restoration. Although decellularized extracellular matrix (dECM) is bioactive, its exhibits inherently dense structures with poor interconnectivity, that severely restrict cellular infiltration and tissue remodeling. To overcome this, we devised an innovative fabrication strategy involving fragmentation, film formation, fiber cutting, and weaving to engineer muscle-derived dECM (mdECM) into woven fiber scaffolds (mFS). This strategy transformed mdECM into tissue-specific “fabrics” with highly interconnected and guidable porosity, directly addressing the limitations of conventional dense mdECM scaffolds (mDS). The mFS dramatically enhanced cell migration in vitro and promoted rapid host-cell infiltration, and vascularization. In a rat VML model, mFS implantation drove aligned myofiber regeneration alongside synchronized vascularization and innervation, leading to significantly superior functional recovery compared to mDS. Single-cell omics reveals a novel mechanism by which lymphatic endothelial cells and regulatory T cells coordinate the inflammatory response to jointly promote muscle regeneration. This work establishes a facile and translatable mdECM tissue fabric platform that integrates the native bioactivity of mdECM with a rationally designed, regenerative-friendly microstructure, thereby providing a transformative strategy for functional reconstruction of complex muscle defects.Keywords: Decellularized extracellular matrix, Fiber scaffolds, Tissue fabrics, Volumetric muscle loss, Muscle regeneration Introduction Skeletal muscle accounts for over 40% of adult body mass and serves as a key tissue responsible for movement, posture maintenance, thermoregulation, and energy metabolism [1]. VML, resulting from trauma (traffic accidents, combat injuries, penetrating trauma, blast injuries) and congenital disorders, often leads to irreversible fibrosis, impaired mobility, and even permanent disability, imposing a heavy burden on patients' quality of life, their families, and socioeconomic systems [2]. Currently, clinical management of VML is limited to autologous muscle transplantation. While this grafting strategy can partially restore muscle structure and function, such a surgical intervention faces considerable challenges including donor site morbidity, flap necrosis and infection, technical complexity, and the necessity for secondary operations [3,4]. In some VML patients, the lack of suitable donor muscles results in persistent defects, which not only cause local functional loss but also trigger systemic metabolic dysregulation, chronic inflammation, and multi-organ complications [5]. Therefore, the effective regeneration and functional repair of VML remains a major unresolved clinical and scientific challenge worldwide. The massive loss of muscle cells and extracellular matrix (ECM) following VML results in the severe disruption of endogenous regenerative microenvironment cues, thereby preventing effective structural and functional reconstruction in situ [3]. The implantation of inductive biomaterials into the defect site to precisely control biophysical and biochemical signals, guide endogenous cell homing, and direct cellular behavior for in situ tissue regeneration has emerged as a key direction to address VML regeneration [6]. Among various candidate materials, mdECM most closely approximates the composition of the natural muscle ECM. It demonstrates advantages over other ECM materials in regulating the immune microenvironment and promoting physiological tissue remodeling in muscle [7]. However, the inherently dense and low-porosity structure of native mdECM severely restricts deep host-cell migration and vascularization, thereby compromising muscle tissue regeneration and resulting in limited functional recovery [6,8]. To overcome this physical barrier, several engineering strategies have been explored to achieve refined structural control over mdECM [9]. For instance, 3D bioprinting can precisely architecture mdECM-based bioinks into ordered porous scaffolds to provide essential microenvironmental cues; however, the stringent rheological requirements of the printing process often lead to a mismatch between structural integrity and appropriate degradation kinetics [10]. Electrospun nanofibers offer critical topological cues to guide the anisotropic alignment of myogenic cells, yet their inherently dense fiber packing and limited pore size frequently hinder deep three-dimensional cell infiltration and host tissue integration [11]. Furthermore, while in vitro self-assembled dECM fibers derived from myogenic cells exhibit exceptional biomimetic anisotropy and successfully guide vascularization and muscle volume restoration, their application is constrained by time-intensive fabrication, technical complexity, low yields, and prohibitive costs [12]. Additionally, although sacrificial microsphere templating can create large-pore dECM hydrogels that promote cell proliferation, the conventional acid- or enzyme-based removal processes risk compromising the endogenous bioactivity of the ECM, and the resulting hydrogels often lack guiding topological structures [13]. Therefore, current approaches frequently struggle to precisely control the porosity of muscle ECM, thereby compromising their ability to effectively promote muscle regeneration [14]. Textile structures inherently form highly interconnected three-dimensional networks through their warp-and-weft fiber assembly [15]. This unique topological structure not only facilitates the transport of oxygen, nutrients, and metabolic waste but also provides physical guidance for cell infiltration and directional growth [15,16]. Fiber-based scaffolds are already widely used in tissue engineering [17]. Previous studies have attempted to fabricate dECM fibers by physically cutting natural membrane materials such as amniotic membrane, small intestinal submucosa, or in vitro-constructed cell sheets, which were then used to prepare patches, artificial blood vessels, and sutures [[18], [19], [20]]. This inspires an innovative strategy: converting mdECM into processable fibers and then reconstructing them through weaving into a mdECM fabric with a highly porous architecture could create a novel class of biomaterials integrating both physical guidance and biochemical induction, thereby offering an improved solution for volumetric tissue regeneration [21]. This study presents an innovative strategy for constructing mdECM-based fabrics, employing an integrated process of fragmentation, film formation, fiber cutting, and weaving to transform mdECM into weavable fibers, which are subsequently layered and assembled into a fibrous scaffold with highly interconnected porosity. We systematically characterized the physicochemical properties of the scaffold and evaluated its performance through in vitro cell experiments, subcutaneous implantation models, and a rat VML defect model, comprehensively assessing its role in promoting cell migration, mediating coordinated regeneration of muscle, blood vessels and nerves, and restoring muscle function. We also analyzed the mechanism of regeneration using single-cell sequencing. The work not only offers a breakthrough solution for the clinical repair of VML, but also establishes a “dECM fabric” strategy that opens a new technical pathway with strong generality and translational potential for the engineering regeneration of other complex tissues and organs. Results Fabrication and physical characterization of mdECM fibers The schematic illustrates the fabrication process of the mdECM fiber, in which porcine muscle tissue was decellularized, homogenized, dried, and sectioned to yield mdECM fibers (Fig. 1A). We performed decellularization using a combined sodium dodecyl sulfate (SDS) and enzymatic digestion strategy, yielding mdECM. H&E staining confirmed the absence of cell nuclei and the retention of abundant ECM components after decellularization. DAPI staining further demonstrated extensive removal of nuclei, and statistical analyses revealed significantly reduced nuclear counts and DNA content in the decellularized muscle tissue (Fig. 1B–D). Proteomic analysis revealed the preservation of key ECM proteins, including collagens, fibronectin, glycoproteins, laminins, and proteoglycans (Supplementary Fig. 1). Preparation and characterization of mdECM fibers. (A)Schematic illustration of the process for preparing mdECM fibers: decellularized muscle was homogenized, fragmented, air-dried, and cut into fibers. (B) H&E and DAPI staining of porcine muscle before and after decellularization. (C, D) Quantitative analysis of nuclear count and DNA content in the materials before and after decellularization (n = 5). (E)Macroscopic images and SEM micrographs of fibers obtained from cutting mdECM films with thicknesses ranging from 100 μm to 500 μm. (F)Width distribution of the different fibers (n = 5). (G) Thickness of the different fibers (n = 5). (H)Tensile stress–strain curves of the fibers. (I)Breaking strength of the different fibers (n = 3). (J)Macroscopic photographs demonstrating the bend (500 μm fiber), knot (500 μm fiber), twist (100 μm fiber), bonding (500 μm fiber), and winding (500 μm fiber) behavior of the mdECM fiber. Bar heights and error bars represent the mean ± s.e.m. Statistical analysis: independent two-tailed t-tests (C, D); one-way ANOVA with Tukey's post-hoc test (F, G, I). ns = not significant, ∗∗∗∗P < 0.0001. Homogenized mdECM of varying quantities was evenly spread on a mesh and air-dried for 24 h to form mdECM membranes approximately 100-500 μm in thickness. Photographs showed that the mdECM membranes were of uniform thickness and exhibited a pale yellow color. These membranes were then cut into fibers using a cutting device. Macroscopic images revealed that the fibers exhibited a uniform width in top-view observations, while scanning electron microscopy (SEM) further demonstrated a relatively dense surface morphology. From a lateral perspective, the fibers maintained a controlled thickness, and SEM imaging indicated a characteristically rough sidewall structure. Quantitative analyses confirmed that the prepared fibers exhibited a uniform width of 0.72 ± 0.04 mm (Fig. 1E and F). Quantitative analysis revealed that the mdECM fibers exhibited relatively uniform thickness, measuring 0.11 ± 0.02 mm, 0.21 ± 0.03 mm, 0.31 ± 0.02 mm, and 0.53 ± 0.02 mm, respectively (Fig. 1G). Tensile mechanical testing showed that breaking strength increased with fiber thickness. Since 100 μm fibers demonstrated the lowest mechanical strength and were prone to fracture, fibers within the 200-500 μm range were chosen for the subsequent experiments (Fig. 1H and I). To evaluate processing adaptability, the fibers were subjected to a series of manipulations including bending into a loop, knotting, twisting, bonding of two fibers together, and winding onto a spool, all of which were performed without structural failure (Fig. 1J). These results demonstrate that the established protocol successfully converts muscle into uniform, mechanically robust, and processable mdECM fibers. Fabrication and physical characterization of mFS The schematic illustrates that mdECM membranes were sectioned into fibers and woven to form mFS (Fig. 2A). Macroscopic images revealed that the mFS was characterized by a crisscrossing arrangement of mdECM fibers, with sparser horizontal fibers and denser vertical fibers. In contrast, the mDS exhibits a relatively denser structure. SEM revealed that the surface and cross-section of the mFS exhibited a porous, interwoven fibrous architecture, with more pronounced porosity as fiber thickness increased. In contrast, mDS exhibited relatively dense surface and cross-sectional morphology (Fig. 2B). The overall thickness of the resulting mFS increased correspondingly with the thickness of the constituent fibers (Fig. 2C). Fabrication and physicochemical characterization of mFS. (A) Schematic illustration of the mFS preparation process: dECM films are cut into fibers and subsequently woven into mFS. (B) Macroscopic and SEM images of mFS and mdECM blocks, captured from both top and lateral views. (C) Thickness measurements of different mFS samples (n = 5). (D, E) Toluidine blue permeability assay was performed on various mFS and mDS samples, followed by quantitative statistical analysis of absorbance (n = 3). (F, G) Tensile stress–strain curves and corresponding fracture strength statistics for different mFS and mDS samples (n = 3). (H) Macroscopic images show that mFS can be fabricated into large-scale constructs and is amenable to stacking, bending, and cutting into letter-shaped forms. Bar chart data are shown as the mean ± s.e.m. Statistical analysis: one-way ANOVA with Tukey's post-hoc test (C, E, G). ns = no significance; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001. Toluidine blue permeation assays confirmed evident permeability in mFS, whereas mDS blocked solution penetration (Fig. 2D). Quantitative analysis further indicated that permeation was enhanced with greater fiber thickness in mFS (Fig. 2E). The porosity of mFS was significantly higher than that of mDS, and the porosity increased with fiber diameters (Supplementary Fig. 2A). Statistical analysis revealed that the lateral pore size of mFS ranged from 139 to 469 μm, while the lateral pore size of mDS was significantly smaller (Supplementary Fig. 2B). Tensile mechanical tests revealed that all mFS samples exhibited lower mechanical strength compared to mDS. Within the mFS group, as fiber thickness increased, the breaking strength gradually improved (Fig. 2F and G). Moreover, mFS shows excellent processability and can be processed into large-scale specimens, such as mFS as long as 10 cm (long pieces), thereby enabling the fabrication of sizable structures. These could be further shaped through stacking of two layers, bending into rings and triangles, or cutting into letter forms (‘NKU’) to fulfill practical processing needs (Fig. 2H). These results indicate that mFS fabricated via weaving possess suitable porous structures and tunable mechanical properties. Cell behaviors regulated by mFS in vitro DID-labeled L6 cells were seeded on mFS and mDS, respectively, to evaluate cytocompatibility and migratory behavior. On day 1 post-seeding, we observed cells inside mFS, and the fluorescence intensity increased over time, indicating continuous inward migration. In contrast, cells on mDS remained confined to the surface, with only a slight increase in fluorescence (Fig. 3A and B). CCK-8 assays showed comparable cell proliferation on both materials, with no significant difference between groups (Fig. 3C). Live/dead staining revealed predominantly green fluorescence (live cells) on both scaffolds, and cell numbers increased noticeably over time, whereas red fluorescence (dead cells) was minimal (Fig. 3D). Quantitative analysis confirmed that cell viability remained above 90% for both mFS and mDS from day 1 to day 5, with no statistical difference between groups (Fig. 3E). F-actin/DAPI staining further demonstrated well-spread cell morphology on both materials. Fluorescence-based quantification indicated increasing cell spreading area over time (Fig. 3F and G). Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis confirmed elevated expression of myogenic differentiation genes (Myo6, Pax7, Myod1, Myf5, Myh9) in mFS compared with mDS (Supplementary Fig. 3). Behavioral changes of L6 cells cultured on mFS and mDS for 1, 3, and 5 days. (A) Distribution of DID-red fluorescently labeled L6 cells on the materials; red indicates DID-labeled cells, black represents the scaffold material. (B) Quantitative analysis of DID fluorescence (n = 3). (C) CCK-8 assay quantifying cell viability (n = 3). (D) Live/dead staining of L6 cells; green represents live cells, red indicates dead cells. (E) Quantitative analysis of cell viability (n = 3). (F) Distribution of L6 cells on mFS and mDS visualized by FITC-phalloidin (green) and DAPI (blue) staining. (G) Quantitative analysis of phalloidin fluorescence intensity (n = 3). Bar chart data are shown as the mean ± s.e.m. Statistical analysis: two-way ANOVA and Tukey's post-hoc test (B, C, E, G), ns = no significance, ∗P < 0.05, ∗∗P < 0.01. Taken together, these results demonstrate that mFS not only supports cell survival and proliferation, but also promotes three-dimensional cell migration. Performance of mFS in vivo mFS and mDS were subcutaneously implanted in rats and harvested at two- and four-weeks post-implantation to evaluate tissue infiltration, vascular ingrowth, and macrophage polarization (Fig. 4A). At two weeks, both mFS and mDS were encapsulated by host tissue. By four weeks, a reduction in volume was observed for both materials (Fig. 4B). Quantitative analysis indicated no significant difference in weight between the two groups (Fig. 4C). DAPI staining revealed extensive nuclear infiltration throughout mFS at both two and four weeks, whereas mDS exhibited fewer nuclei compared to mFS (Fig. 4D). Quantification confirmed that the number of infiltrated cells in mFS was significantly higher than in mDS (Fig. 4E). H&E staining showed that cells and ECM had populated most of the mFS by two weeks, while mDS still contained regions devoid of cellular infiltration. At four weeks, mFS degradation accompanied by volume shrinkage was observed, whereas mDS retained residual material with persistent non-infiltrated areas. Quantitative analysis demonstrated that the percentage of non-infiltrated area in mDS was significantly larger than that in mFS (Fig. 4F–I). Tissue responses to subcutaneously implanted mFS and mDS in rats at 2 and 4 weeks. (A) Schematic of the subcutaneous implantation of scaffolds and sample collection at different time points. (B, C) Macroscopic views and weight measurements of explanted scaffolds at 2 and 4 weeks (n = 3). (D, E) DAPI staining showing cell infiltration into the implants and corresponding quantification (n = 3). (F–I) H&E staining illustrating tissue ingrowth into the implants and quantitative analysis of the ingrown area (n = 3). (J, K) H&E staining demonstrating microvessel ingrowth and associated quantification (n = 3). (L, M) α-SMA immunofluorescence (green) indicating functional vascular smooth muscle layers and corresponding quantitative analysis (n = 3). (N–Q) iNOS (red) and CD206 (green) staining showing the distribution of pro-inflammatory (iNOS+) and pro-regenerative (CD206+) macrophages, with corresponding quantification (n = 3). Bar chart data are shown as the mean ± s.e.m. Statistical analysis: independent two-tailed t-tests (G, I), two-way ANOVA and Tukey's post-hoc test (C, E, K, M, O, Q), ns = no significance, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. Vascularization was assessed next. H&E staining indicated that mFS supported extensive microvessel formation at two weeks, which showed a tendency to increase over time. In contrast, mDS displayed minimal vascularization at two weeks and only limited capillaries at four weeks. Quantitative results confirmed that the number of vessels in mFS was significantly higher than in mDS at both time points (Fig. 4J and K). Immunostaining for α-smooth muscle actin (α-SMA), a marker of pericytes indicative of mature capillaries, revealed neovascularization within mFS at 2 weeks, which was absent in mDS. At 4 weeks, mFS demonstrated an increase in microvessels, with the number significantly greater than that in the mDS group. Quantitative analysis confirmed these observations (Fig. 4L and M). Macrophage infiltration was subsequently evaluated. Immunostaining for iNOS+ macrophages revealed them to be predominantly localized within the mFS, while in the mDS group, they were mainly restricted to the periphery. At two weeks, the numbers of iNOS+ macrophages were significantly higher in mFS than in mDS. This difference had diminished by four weeks (Fig. 4N and O). In both two- and four-week samples, abundant CD206+ macrophages were distributed throughout mFS, whereas few were detected inside mDS. Quantification confirmed significantly higher CD206+ macrophage counts in mFS compared to mDS (Fig. 4P and Q). These results demonstrate that mFS, relative to mDS, enhances cellular infiltration, promotes neovascularization, and recruits more CD206+ macrophages. mFS promotes muscle regeneration in a rat VML model mFS (measuring 1.0 × 0.5 × 0.5 cm) and mDS were implanted into a rat tibialis anterior VML model for 2 and 8 weeks to evaluate their ability to promote in situ muscle regeneration (Fig. 5A). At two weeks, both mFS and mDS were encapsulated by host tissue. The untreated group exhibited muscle atrophy, while the autograft group showed better integration between the graft and surrounding muscle. By 8 weeks, both mFS and mDS had completely degraded, with no obvious boundary between the wound and adjacent muscle tissue (Fig. 5B). The mass ratio (injured side/normal side) measured at 8 weeks revealed that the mFS group was significantly higher than the mDS and untreated groups, though still lower than the autograft group (Fig. 5C). H&E staining at two weeks showed substantial host cell infiltration into both mDS and mFS. The untreated group displayed atrophic characteristics, whereas the autograft graft region was heavily infiltrated by cells. By 8 weeks, mFS and mDS had fully degraded. The mFS and autograft groups exhibited abundant regenerated muscle fibers, while the mDS and untreated groups still showed noticeable depressions at the defect site (Fig. 5D). Quantification of DAPI-stained nuclei indicated rapid cell infiltration in the mFS and autograft groups at two weeks, which was significantly greater than in the mDS group. By 8 weeks, the mDS group displayed a higher number of cells in the defect area compared to the mFS group (Fig. 5E, Supplementary Fig. 4). Masson's trichrome staining revealed substantial collagen deposition at the implantation site in both mFS and mDS groups at two weeks, as well as in the periphery of the autograft transplant. Quantification confirmed that collagen deposition in the mFS group was significantly higher than in the autograft and untreated groups. At 8 weeks, the mFS and autograft groups showed minimal collagen, whereas the mDS group still exhibited abundant collagen in the defect area. The mDS group had significantly more collagen than the mFS and untreated groups (Fig. 5F and G). Muscle regeneration in rat defects implanted with mFS and mDS. (A) Schematic of the scaffold implantation in rat tibialis anterior muscle and experimental timeline. (B) Macroscopic images of implants and harvested tissues at different time points. (C) Ratio of muscle mass (defect side/normal side) at 8 weeks post-implantation (n = 3). (D) H&E staining showing tissue ingrowth. (E) Quantification of nuclear counts within scaffolds over time (n = 3). (F, G) Masson's trichrome staining revealing collagen deposition and its quantitative analysis (n = 3). (H, I) Pax7 (green) immunofluorescence showing distribution of muscle satellite cells and corresponding quantification (n = 3). (J, K) Desmin (green) immunofluorescence indicating myofiber formation and its quantitative evaluation (n = 3). Bar chart data are shown as the mean ± s.e.m. Statistical analysis: one-way ANOVA and Tukey's post-hoc test (C), two-way ANOVA and Tukey's post-hoc test (E, G, I, K), ns = no significance, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001. We further evaluated muscle satellite cells and myofiber regeneration using immunofluorescence staining. Pax7 staining at two weeks showed that satellite cells were predominantly located within the mFS scaffold and the autograft region, with fewer cells in the mDS group. Quantification data confirmed that the mFS group had significantly more Pax7+ cells than the mDS and untreated groups (Fig. 5H and I). Desmin immunostaining at two weeks revealed the presence of fine myofibers only in the mFS and autograft groups, with almost none in the mDS group. The autograft group had a significantly larger desmin-positive area than the mDS and untreated groups. By 8 weeks, the mFS and autograft groups showed extensive nascent myofibers, whereas myofibers in the untreated and mDS groups remained thin. The desmin-positive area in the mFS group was significantly larger than in the untreated and mDS groups (Fig. 5J and K). Together, these results demonstrate that mFS facilitate rapid host cell infiltration and promote muscle-fiber regeneration during recovery from VML in rat models. mFS promotes vascularization, innervation, and functional recovery in a rat VML model Vascularization and innervation are key indicators of functional muscle regeneration. We evaluated vascularization using α-SMA staining. At 2 weeks, abundant functional capillary had formed in both the mFS and autograft groups. In contrast, less functional capillary ingrowth was observed within the scaffolds in the mDS and untreated groups. By 8 weeks, the number of functional capillaries had decreased. However, in the mFS group, numerous functional capillaries remained distributed throughout the defect site(Fig. 6A). Quantitative analysis confirmed that the mFS and autograft groups exhibited significantly higher functional capillary densities than the mDS and untreated groups at 2 weeks. No significant differences in functional capillary numbers were detected among groups at 8 weeks (Fig. 6B). Vascular regeneration, neural regeneration, macrophage polarization, and muscle force recovery in rat VML defects implanted with mFS and mDS. (A, B) Immunofluorescence staining of α-SMA (green) showing functional capillary ingrowth and quantitative analysis (n = 3). (C, D) Representative CMAP recordings and maximum amplitude statistics at 8 weeks (n = 3). (E, F) NF09 (green) immunofluorescence staining showing neural distribution and quantitative analysis (n = 3). (G, H) Representative tetanic force measurement curves and quantitative analysis (n = 3). (I, J) Immunofluorescence staining and quantification (n = 3) for CD206 (green), showing the distribution of pro-regenerative macrophages at 2 and 8 weeks. (K, L) Immunofluorescence staining and quantification (n = 3) for iNOS (red), indicating the distribution of pro-inflammatory macrophages. Bar chart data are shown as the mean ± s.e.m. Statistical analysis: one-way ANOVA and Tukey's post-hoc test (D, H), two-way ANOVA and Tukey's post-hoc test (B, F, J, L), ns = no significance, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001. Compound muscle action potential (CMAP) recordings performed at 8 weeks showed that all groups responded to electrical stimulation (Fig. 6C). Quantitative analysis revealed that CMAP amplitudes were significantly greater in the mFS group than in the mDS and untreated groups (Fig. 6D). Immunofluorescence staining for neurofilament (NF09) revealed minimal NF09-positive expression at the injury site across all groups at 2 weeks. By 8 weeks, quantitative analysis indicated detectable NF09-positive expression in the mFS and autograft groups, though levels in both remained lower than in the normal group (Fig. 6E and F). Tetanic force measurements at 8 weeks demonstrated that all groups generated force in response to electrical stimulation. The mFS group exhibited significantly greater tetanic force compared to the mDS group (Fig. 6G and H). Together, these results indicate that the mFS group promoted vascularization and innervation during muscle repair and enhanced the recovery of muscle strength relative to the mDS group. Finally, we evaluated the effect of the scaffolds on the abundance and distribution of both pro-inflammatory and pro-regenerative macrophages. The CD206 marker is an indicator for pro-regenerative macrophages. Immunofluorescence staining for CD206 revealed that at 2 weeks, numerous pro-regenerative macrophages were distributed within the mFS and around the muscles in the autograft group, whereas minimal CD206-positive signal was detected inside the mDS. Quantitative analysis revealed that the mFS group contained a significantly greater number of CD206-positive cells than the mDS and untreated groups at 2 weeks. A marked reduction in these cells was observed across all groups by 8 weeks (Fig. 6I and J). The iNOS markers are indicators of pro-inflammatory macrophages. Immunofluorescence staining for iNOS showed that at 2 weeks, pro-inflammatory macrophages were predominantly localized in the mDS group, with quantitative results indicating a significantly greater number compared to the normal group; no significant differences were observed among groups at 8 weeks (Fig. 6K and L). These findings demonstrate that mFS, compared to mDS, more effectively promotes macrophage polarization toward a pro-regenerative phenotype and fosters a superior pro-regenerative microenvironment during rat muscle regeneration. A single-cell atlas of the rat VML defects implanted with mFS and mDS To comprehensively elucidate the mechanism underlying mFS-mediated muscle regeneration, we prepared single-cells suspensions from the explants of mFS and mDS at 2 weeks for single-cell RNA sequencing. Based on the known conserved marker gene expression profiles, 10 major cell types were annotated and visualized using t-distributed stochastic neighbor embedding (tSNE, Fig. 7A). The mFS group exhibited higher proportions of muscle cells, whereas the mDS group showed increased proportions of macrophages. Both groups also contained endothelial cells, B cells, fibroblasts, neutrophils, pericytes, lymphatic endothelial cells, T cells, and plasmacytoid dendritic cells (Fig. 7B). A single-cell atlas of the rat VML defects implanted with mFS and mDS. (A, D, Q) tSNE dimensional reduction representing cells categorized into several main clusters with each cell color-coded based on its cell type. (B, E, J, L, R, W) Bar plots showing cell proportions of each cluster presented by percentages of sequenced cells. (C, I, K, P, V) Dot plot of cells showing the expression of canonical markers of each cell type. Dot size corresponds to the proportion of cells within the group expressing each gene, and color corresponds to expression level. (F, M, S, X) Dot plot of the enriched GO of highly expressed genes in each subpopulation. Dot size corresponds to the count of cells within the group expressing each gene, and dot color corresponds to GO enrichment p-value. (G, H, N, O, T, U, Y, Z) The differentiation trajectory of cells, with each colour coded for subpopulations and pseudotime, and density plots showing the number of pseudobulked cells in the relative pseudotime scale. Macrophages were subclustered into Macro_Cd206+ macrophages (macrophages expressing Cd206), Macro_ Nrp1+ (macrophages expressing Nrp1) macrophages, Macro_ Inos+ (macrophages expressing Inos) macrophages, Macro_Spp1+ (macrophages expressing Spp1) macrophages, and Macro_Proliferating macrophages (macrophages expressing Mki67) subpopulations (Fig. 7C) and visualized via tSNE (Fig. 7D). Statistical analysis demonstrated that the mFS group had a higher proportion of Macro_Cd206+ and Macro_ Nrp1+ macrophages than the mDS group (Fig. 7E). Gene Ontology (GO) enrichment analysis revealed that Macro_Cd206+ macrophages were enriched in pathways related to the immune system process and the others (Fig. 7F). To delineate the transition states associated with different types of macrophages, pseudotemporal analysis was performed. Macro_ Inos+ macrophages dominated early stages, while Macro_ Nrp1+ macrophages and Macro_Cd206+ macrophages appeared at later stages in the transition states. Compared with the mDS group, the mFS group appeared at later stages in the transition states, characterized primarily by Macro_ Nrp1+ macrophages and Macro_Cd206+ macrophages (Fig. 7G and H). Further subclustering of T cells identified T_Cd4+ (Cd4) and T_Cd8+(Cd8) subpopulations (Fig. 7I). Quantitative analysis revealed a higher proportion of T_Cd4+ cells in the mFS group compared to the mDS group (Fig. 7J). T_Cd4+ cells were subclustered into T_Foxp3+ cells (Foxp3, Ctla4, Lag3) and T_Tcf7+ (Tcf7, Ccr7) cells subpopulations (Fig. 7K). Quantitative analysis revealed a higher proportion of T_Foxp3+ cells in the mFS group compared to the mDS group (Fig. 7L). GO enrichment analysis indicated that T_Foxp3+ cells were predominantly enriched in pathways related to wound healing, tissue remodeling, and others (Fig. 7M). Pseudotime analysis was performed on T cells. T_Tcf7+ cells dominated middle stages, while T_Foxp3+ cells remain abundant at later stages in the transition states. Compared with the mDS group, the mFS group appeared at later stages in the transition states, characterized primarily by T_Foxp3+ cells (Fig. 7N and O). Endothelial cells were subclustered into Endo_Cdh5+_Vwf+ (Cdh5, Vwf) endothelial cells, Endo_Kdr+_Flt1+ (Kdr, Flt1, Jag, Mki67, Hey1) endothelial cells subpopulations (Fig. 7P) and visualized via tSNE (Fig. 7Q). Statistical analysis demonstrated that the mFS group had a higher proportion of proliferative activation Endo_Kdr+_Flt1+ endothelial cells than the mDS group (Fig. 7R). GO enrichment analysis revealed that endothelial cells were enriched in pathways related to the vasculature development and the others (Fig. 7S). To delineate the transition states associated with different types of endothelial cells, pseudotemporal analysis was performed. Endo_Cdh5+_Vwf+ endothelial cells dominated early stages, while Endo_Kdr+_Flt1+ endothelial cells appeared at later stages in the transition states. Compared with the mDS group, the mFS group appeared at later stages in the transition states, characterized primarily by Endo_Kdr+_Flt1+ endothelial cells (Fig. 7T and U). Lymphatic endothelial cells were subclustered into two subpopulations: LECs_Lyvel+_Ccl21+ lymphatic endothelial cells (Prox1, Pdpn, Flt4, Cdh5, Lyve1, Ccl21, Nrp2, Itag9) and LECs_Mki67+_Top2a+ lymphatic endothelial cells (Mki67, Top2a) (Fig. 7V). Statistical analysis revealed that the proportion of proliferative activation LECs_Mki67+_Top2a+ lymphatic endothelial cells in the mDS group lower than the mFS group (Fig. 7W). GO enrichment analysis indicated that lymphatic endothelial cells were enriched in pathways related to chemotaxis, inflammatory response and others (Fig. 7X). Pseudotime analysis was performed on lymphatic endothelial cells. LECs_Mki67+_Top2a+ lymphatic endothelial cells dominated later stages in the transition states. Compared with the mDS group, the mFS group appeared at later stages in the transition states, characterized primarily by LECs_Mki67+_Top2a+ lymphatic endothelial cells (Fig. 7Y and Z). Altered cell-cell signaling interactions in rat VML defects implanted with mFS and mDS Analysis of intercellular communication and interactions was performed using cellchat. Based on receptor and ligand expression, the overall strength of incoming/outgoing communication in cell subpopulations showed that compared with the mDS group, the mFS group exhibited increased interaction strength, encompassing both outgoing and incoming signals in LECs_Mki67+_Top2a+ cells and increased incoming signals in muscle cells and T_Foxp3+cells (Fig. 8A and B). Heatmap analysis of cell-cell interactions revealed increased interactions between the LECs_Mki67+_Top2a+ subpopulation itself, between LECs_Mki67+_Top2a+ and muscle cells, between LECs_Mki67+_Top2a+ and T_Foxp3+ cells, and between T_Foxp3+ cells and muscle cells in the mFS group compared to the mDS group (Fig. 8C). The chord plot revealed that compared to the mDS group, cell communication from both lymphatic endothelial cells and T cells to muscle cells was significantly enhanced in the mFS group. More specifically, cell communication from LECs_Mki67+_Top2a+ to both T_Foxp3+ and muscle cells, and from T_Foxp3+ to muscle cells was significantly enhanced (Fig. 8D). Heatmap analysis of signaling patterns revealed a marked increase in interactions involving the LECs_Mki67+_Top2a+ subpopulation within the mFS group (Fig. 8E). Altered cell-cell signaling interactions in rat VML defects implanted with mFS and mDS. (A) The overall strength of incoming and outgoing communication, as determined by the expression levels of ligands and receptors. (B) Two-dimensional visualization representing the overall strength of incoming/outgoing communication in cell subpopulation based on receptor and ligand expression, with dot size proportional to the count of cells in each cluster. (C) Heatmap depicting the strength of interactions among cell subpopulations. Darker blue hues represent a reduction in interaction strength in the mFS relative to the mDS, while darker red hues represent an increase relative to the mDS. (D) Chord diagram of cellular communication relationship among cell clusters. The connecting lines represent potential cellular communication between cell types, with line thickness proportional to the strength of ligand-receptor pairs expressed within the connected cell types. Darker blue hues denote a reduction in the strength of ligand-receptor pairs in the mFS relative to the mDS, while darker red hues denote an increase relative to the mDS. (E) Identify the pathways that contribute most significantly to the outgoing or incoming signaling of specific cell populations. The color intensity of each cell represents the magnitude of signaling strength. (F, G) Dot plot depicting expression of hypoxia-associated genes and endothelial growth-associated genes in the mFS and mDS. (H) Comparison of the information flow between LECs_Mki67+_Top2a+ and muscle cells. (I) Dot plot illustrating the upregulated signaling ligand-receptor pairs between cell clusters in mFS versus mDS. (J) Dot plot depicting expression of regeneration-associated genes in LECs_Mki67+_Top2a+ within the mFS and mDS. (K) Comparison of the information flow between LECs_Mki67+_Top2a+ and neutrophil. (L) Dot plot illustrating the upregulated signaling ligand-receptor pairs between cell clusters in mFS versus mDS. (M) Dot plot depicting expression of neutrophil chemotaxis-associated genes in LECs_Mki67+_Top2a+ within the mFS and mDS. (N) Comparison of the information flow between LECs_Mki67+_Top2a+ and T cells. (O) Dot plot illustrating the upregulated signaling ligand-receptor pairs between cell clusters in mFS versus mDS. (P) Dot plot depicting expression of T cells chemotaxis-associated genes in LECs_Mki67+_Top2a+ within the mFS and mDS. (Q) Comparison of the information flow between T_Foxp3+ to muscle cells. (R) Dot plot illustrating the upregulated signaling ligand-receptor pairs between cell clusters in mFS versus mDS. (S) Dot plot depicting expression of regeneration-associated genes in T_Foxp3+ within the mFS and mDS. (T) Schematic diagram depicting the mechanism of mFS induces muscle regeneration. The aforementioned cellchat analysis identified the predominant cell types exhibiting increased interactions; subsequently, a detailed investigation of the cellchat analysis between these primary cell populations was performed. Lymphatic endothelial cells function upstream in cellular interactions, and their growth may be regulated by the scaffold's interstitial space and growth factors. Compared with the mFS group, the mDS group exhibited higher expression of genes associated with hypoxia (Hif1a, Sod2), indicating poor porosity and impaired fluid mobility, which further hindered lymphangiogenic sprouting and elongation (Fig. 8F). Meanwhile, the mFS group exhibited higher expression of genes associated with endothelial growth factor (Vegfd, Vegfb, Vegfc) (Fig. 8G). LECs_Mki67+_Top2a+ exhibited increased interactions with multiple cell types. Information flow analysis revealed that between LECs_Mki67+_Top2a+ and muscle cells, regeneration-related genes (Apoe, Ptn, Csf, Laminin) were upregulated in the mFS group (Fig. 8H). We compared receptor and ligand expression among both cell subpopulations. Relative to the mDS group, expression of the Lama4-Cd44 ligand-receptor pair from LECs_Mki67+_Top2a+ to muscle cells was elevated in the mFS group (Fig. 8I). Furthermore, compared with the mDS group, LECs_Mki67+_Top2a+ in the mFS group exhibited upregulated expression of genes associated with growth factors (Igf1, Hgf, Apoe), indicating a pro-proliferative environment for muscle cells (Fig. 8J). In addition to muscle cells, LECs_Mki67+_Top2a+ also exhibited significant interactions with neutrophils and T cells. Information flow analysis revealed that among LECs_Mki67+_Top2a+, T cells, and neutrophil, genes associated with inflammatory regulation (Fn1, App) were upregulated in the mFS group (Fig. 8K and N). Relative to the mDS group, expression of the Fn1_(Itgav + Itgb3) and App_Cd74 ligand-receptor pair from LECs_Mki67+_Top2a+ to neutrophil as well as from LECs_Mki67+_Top2a+ to T cells was elevated in the mFS group (Fig. 8L and O). Furthermore, compared with the mDS group, LECs_Mki67+_Top2a+ in the mFS group exhibited higher expression of genes associated with neutrophil and T cell chemotaxis (Cxcl12, Cxcl1, Cxcl9), suggesting that LECs_Mki67+_Top2a+ recruit neutrophils and T cells to mitigate the excessive accumulation of inflammatory cells (Fig. 8M and P). In addition to LECs_Mki67+_Top2a+, T cells also exhibited significant interactions with muscle cells. Information flow analysis revealed that between T_Foxp3+ and muscle cells, genes associated regeneration-related genes (Apoe, Igf) were upregulated in the mFS group (Fig. 8Q). Relative to the mDS group, expression of the Apoe_(Trem2+Tyrobp) ligand-receptor pair from T_Foxp3+ to muscle cells was elevated in the mFS group (Fig. 8R). The dot plot also revealed a high expression of Apoe (Fig. 8S). We validated the relevant cells and pathways using immunofluorescence staining and RT-qPCR. Hif1a is highly expressed under hypoxic conditions. At 2 weeks post-injury, the proportion of Hif1a+ cell was significantly higher in the mDS group compared to the mFS group (Supplementary Fig. 5A). Top2a, a marker gene for the LECs_Mki67+_Top2a+ cluster, exhibited a significantly lower number of positive cells in the mDS group than in the mFS group. Similarly, the expression of Foxp3, a marker for T_Foxp3+ cells, was significantly reduced in the mDS group relative to the mFS group. Finally, the expression of the muscle cells marker Pax7 was significantly higher in the mFS group than in the mDS group (Supplementary Fig. 5B). Based on these analyses, we elucidated the mechanism whereby mFS promotes muscle regeneration. Compared with the mDS group, LECs_Mki67+_Top2a+ in the mFS group alleviated inflammation by modulating the inflammatory response and recruitment of neutrophils and T cells; furthermore, it acted synergistically with T_Foxp3+ to enhance the pro-regenerative effects on muscle cells (Fig. 8T). Discussion To address the clinical challenge of VML regeneration, this study employed a weaving strategy to reconstruct tissue-specific mdECM, successfully developing a “mdECM fabric” that combines inherent bioactivity with a regenerative porous architecture. Moving beyond the limitations of conventional mdECM materials that act merely as passive fillers, this system not only facilitates rapid cell migration and infiltration both in vitro and in vivo, but also promotes coordinated regeneration of muscle, vasculature, and nerves in a rat VML model, ultimately leading to significant functional muscle recovery. mdECM materials are widely employed in biomedical research such as muscle defect repair due to their rich bioactive components and immunomodulatory properties [14]. However, current engineering strategies for mdECM still face significant challenges in balancing structure, bioactivity, and mechanical properties [22]. While direct implantation of mdECM materials preserves their biochemical activity well, their naturally dense matrix severely limits deep host-cell infiltration and vascularization [23]. Injectable hydrogels can serve as carriers for growth factors such as insulin-like growth factor-1 and enhance regenerative signaling, but their rapid degradation frequently fails to match the spatiotemporal process of long-term muscle remodeling [24]. Although 3D printing enables precise control over scaffold porosity, it often struggles to simultaneously maintain printability of bioinks, preserve sufficient biological activity, and achieve adequate mechanical strength [10]. In summary, existing methods are limited by the inherent processing characteristics of mdECM, making it difficult to construct a regenerative microenvironment that combines ideal mechanical properties with a highly interconnected porous structure while fully retaining native bioactivity [25]. To address this bottleneck, we draw inspiration from textile weaving principles and establish an innovative green fabrication route involving physical fragmentation, film formation, fiber cutting, and weaving to produce fabrics with a highly interconnected porous architecture. This transformation into a three-dimensional network structure forms the physical foundation for the outstanding pro-regenerative performance of the mFS. Compared with existing technologies, our strategy offers distinct advantages in terms of environmentally benign, controllable, and clinically translatable [26]. Unlike methods that rely on solvents or high temperatures, our “dry” weaving process is gentler and maximizes the preservation of the native bioactivity of mdECM, thereby promoting satellite cell recruitment, proliferation, and differentiation to accelerate muscle repair [27]. More importantly, weaving is a highly mature industrial technology in which pore size, fiber orientation, fabric thickness, and mechanical properties can be finely tuned by adjusting fiber diameter, weaving density, and pattern design [15]. This provides unprecedented flexibility for future development of personalized scaffold products tailored to different types of tissue defects. A highly interconnected porous network is essential for driving rapid cellular infiltration and integration with host tissue, while also providing the necessary conditions for establishing a pro-regenerative immune microenvironment and achieving effective muscle regeneration [[28], [29], [30]]. In the context of VML repair, particularly with the implantation of large-scale scaffolds, the ability to achieve rapid vascularization and sufficient cellular infiltration is critical. Large implants often suffer from inadequate nutrient delivery and metabolic waste diffusion in their central regions, which can lead to localized necrosis, exacerbate inflammatory responses, disrupt immune homeostasis, and ultimately result in implant failure [31,32]. The mFS developed in this study features fully interconnected porous architecture. In vitro experiments demonstrated that cells could distribute throughout the entire scaffold within 24 h. In rat subcutaneous implantation and muscle defect repair models, centimeter-scale mFSs exhibited significant neovascularization and complete cellular infiltration as early as two weeks post-implantation. This structure facilitates the rapid infiltration of myogenic progenitor cells into the scaffold interior during the early stages of repair, as corroborated by the robust presence of nascent myofibers at 2 weeks. Furthermore, these scaffolds provide essential structural conduits for synchronized angiogenesis and lymphangiogenesis. The rapid establishment of a functional microvascular network ensures sufficient oxygen and nutrient supply for metabolically demanding regenerating muscle. Furthermore, the development of lymphatic vessels within the mFS group plays a crucial role in immune clearance and tissue fluid homeostasis, creating a supportive niche that rivals the natural vascularized environment of an autograft [33,34]. Our multi-scale analysis reveals that the mFS architecture actively modulates the inflammatory landscape. Compared to the mDS group, the mFS group exhibits a significantly higher recruitment of Cd206+ macrophages and Foxp3+ regulatory T cells [35,36]. Our cell-cell communication analysis further suggests a synergistic axis where lymphatic endothelial cells and immune cells co-regulate the inflammatory resolution, effectively shifting the microenvironment from chronic inflammation to constructive remodeling [37]. Certainly, this study has limitations that warrant further investigation. First, the scale of defects and biomechanical environment in the rat VML model differ from those in human clinical scenarios, and validation in large-animal models (pig or canine) will be essential for clinical translation. Second, enhancing the mechanical strength of the material would enable the fabrication of larger constructs and more complex weaving processes, thereby expanding its potential for repairing tissue defects. Finally, athough the mFS group significantly outperforms the mDS group, its functional recovery (muscle force and reinnervation) approaches but does not fully match the levels achieved by autografts. This is understandable given that autografts provide an immediate, intact niche comprising mature myofibers, a pre-existing vascular network, and native neurotrophic cues that are challenging to fully replicate in any synthetic or bioengineered scaffold [[38], [39], [40]]. In summary, we have developed an innovative mdECM-based textile platform. This work not only provides an efficient, translatable, and functionally robust biomaterial for VML repair, but also introduces a “textile-inspired biomaterial design” concept that opens a promising new technological avenue for regenerating or engineering other vascularized and structurally complex solid organs, such as the liver and kidney. Conclusion We developed tissue fabrics, novel and readily processable textile materials derived from tissue-specific mdECM. These fabrics were fabricated by homogenizing decellularized tissues, spreading and drying the resulting slurry into membranes, and then cutting them into mdECM fibers that were woven into a scaffold. Using mdECM as a prototype, we demonstrated that the resulting mFS possessed a comprehensively interconnected porous structure and mechanical strength suitable for bending, weaving, and secondary processing such as cutting to refined shapes. In vitro, the mFS effectively promoted migration, proliferation, and viability of L6 cells. In a rat subcutaneous implantation model, mFS facilitated rapid cell infiltration and recruited pro-regenerative macrophages. Moreover, in a rat tibialis anterior muscle defect model, the mFS significantly enhanced muscle cell ingrowth, vascularization, innervation, and functional force recovery. Single-cell omics unveils a novel mechanism whereby lymphatic endothelial cells and regulatory T cells orchestrate the inflammatory response to cooperatively promote muscle regeneration. These preclinical findings collectively underscore the promising potential of the mFS for future clinical applications. Methods Animal experiments All animal procedures were approved by the Animal Ethics Committee of Nankai University, Tianjin (Approval No. 22025-SYDWLL-000495) and conducted in accordance with the Guide for the Care and Use of Laboratory Animals. A total of 42 male Sprague Dawley rats aged 8-10 weeks (weight range: 280-320 g) were obtained from SPF Biotechnology Co., Ltd. (Beijing, China). Rats were randomly housed in cages (three animals per cage) with ambient temperature maintained at 23 ± 3 °C, relative humidity at 40-70%, and a 12-h light/dark cycle. Standard food and water were provided ad libitum. Before surgery, rats were fasted for 12 h with free access to water. Postoperatively, all animals were placed on heating pads to maintain stable respiration and body temperature. Twelve rats were used for subcutaneous implantation experiments, and 30 rats were used for the VML model. In all experiments, data were obtained with a sample size of n = 3 per group. Materials Porcine muscle tissue was procured from Tianjin Ershang Yingbin Meat Food Co., Ltd. (Tianjin, China). Deoxyribonuclease I (DNase I), ribonuclease (RNase), 4% paraformaldehyde solution, and sodium dodecyl sulfate (SDS) were obtained from Solarbio (Beijing, China). 3-0 and 9-0 needle sutures were sourced from Ningbo Medical Needle Co., Ltd. (Zhejiang, China). Hematoxylin and eosin (H&E, DH006; Leagene) and Masson's trichrome stain (DC0032; Leagene) were purchased from Leagene. Fabrication of mFS Porcine muscle tissues were sliced into thin sections and sterilized with 0.5% peracetic acid. The sterilized muscle was then eluted with 1% SDS for 3 days, followed by rinsing with PBS solution for 5 days. Subsequently, the samples were treated with DNase and RNase solutions at 37 °C for 24 h and washed again with PBS for 5 days. The decellularized muscle was blended with a certain proportion of water using a homogenizer, spread on a filter mesh, and air-dried at room temperature for 24h to obtain mdECM membranes. Regarding homogenization, specifically, first weigh the wet decellularized matrix, add deionized water in an amount three times its weight, and place the mixture into a homogenizer (SUPOR, JP37V-1300, Zhejiang Supor Co., Ltd., China). Set the power to gear 5 and homogenize for 1 min. These membranes were cut into standardized wide strips using an equidistant array of blades, yielding mdECM fibers. The fibers were woven into a scaffold in a crisscross pattern using a custom-built laboratory device, producing the final mFS. Directly decellularized muscle materials served as the control group. Characterization of mdECM Muscle tissues before and after decellularization were subjected to H&E and DAPI staining to assess nuclear removal. Residual DNA content in the samples was quantified using the Quant-iT PicoGreen dsDNA Assay (Invitrogen, Grand Island, NY) according to the manufacturer's instructions. Proteomic analysis of muscle and decellularized muscle was performed using liquid chromatography-tandem mass spectrometry (LC-MS/MS) (Gene Denovo Biotechnology, Guangzhou, China). The acquired mass spectrometry data were processed with MaxQuant software (version 1.5) and searched against the NCBInr database for protein identification. Mechanical characterization of scaffolds The fracture strength of mFS and mDS was measured using a universal testing machine (HY0580, Shanghai Hengyi Precision Instrument Co., Ltd., China). Samples of mFS and mDS were cut into 1 × 2 cm sheets for testing, and mechanical curves were recorded. Physical characterization of scaffolds The macro- and microstructures of mFS and mDS were examined using stereomicroscopy and scanning electron microscopy (SEM, ThermoFisher Scientific Europe NanoPort, Eindhoven, Netherlands). A toluidine blue penetration assay was performed to assess solution permeation through the material placed between two glass bottles; absorbance in the lower bottle was measured for statistical analysis. Scaffold porosity was measured via the ethanol displacement method. Cellular behavior assays of scaffolds Cell Migration Assay: L6 cells labeled with DID (D4019, UElandy Inc, China) were seeded on 1.0 × 1.0 × 1.0 cm mFS and mDS at a density of 1.0 × 105 cells/mL. Cultures were maintained in medium consisting of 90% Dulbecco's Modified Eagle Medium (DMEM) and 10% FBS, with medium replaced daily. After 1, 3, and 5 days of culture, cells were fixed with paraformaldehyde and sectioned using a cryostat (CM1950, LEICA, Germany). Sectioned samples were automatically scanned with a fluorescence microscope (Thunder, Leica, Germany). Fluorescence intensity was quantified using ImageJ software v1.5. Cell Proliferation Assay: L6 cells were seeded on 1.0 × 1.0 cm mFS and mDS at a density of 1.0 × 104 cells/mL. After 1, 3, and 5 days of culture, cell proliferation was assessed by measuring the absorbance in CCK-8 assays. Cell Viability and Adhesion Assay: L6 cells were seeded on 1.0 × 1.0 cm mFS and mDS at a density of 1.0 × 104 cells/mL. After 1, 3, and 5 days of culture, live/dead staining (Solarbio, China) and phalloidin/DAPI (Sigma-Aldrich) staining were performed. Images were captured using a confocal laser scanning microscope (TSC SP8, Leica, Germany). Cell differentiation assay: To evaluate the regulatory impact of mFS on gene expression in L6 cells, RT-qPCR was performed. Total RNA was extracted from cells following a 5-day culture period, with concentrations quantified using a NanoDrop spectrophotometer (NanoDrop Technologies, USA). Real-time PCR was executed via the SYBR Green detection method on a CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Relative mRNA expression levels were calculated using the 2−ΔΔCT method, normalized to the β-actin. Detailed primer sequences are available in Supplementary Table 1. Subcutaneous implantation experiment mFS and mDS (1.0 × 1.0 × 1.0 cm) were implanted subcutaneously into the dorsal region of rats. Samples were harvested at 2- and 4-weeks post-implantation, fixed with paraformaldehyde, and sectioned using a cryostat. Infiltration was assessed via DAPI and H&E staining. Immunofluorescence staining was performed with the following primary antibodies: iNOS (1:200, Abcam, ab15323) for pro-inflammatory macrophages, CD206 (1:200, Abcam, ab64693) for pro-regenerative macrophages, and α-SMA (1:100, Abcam, ab7817) for blood vessels. Sections were permeabilized with 1% Triton X-100, incubated overnight with primary antibodies in goat serum, washed with PBS, incubated with secondary antibodies, and mounted with DAPI. Images were acquired using a confocal laser scanning microscope (TSC SP8, Leica, Germany). Rat muscle defect model Rats were randomly assigned to four groups: untreated, mFS, mDS, and autograft. Anesthesia was induced with ketamine (40 mg/kg)–xylazine (5 mg/kg)–acepromazine (1 mg/kg) and maintained with isoflurane inhalation throughout the procedure. For the tibialis anterior muscle defect, hair was removed from the leg, the skin was incised, and a 10 × 5 × 5 mm muscle segment was excised to create a defect representing approximately 40% of the muscle volume. The implanted material was secured with sutures at four corners. All surgeries were performed under specific pathogen-free (SPF) conditions. Finally, rats were euthanized at 2- and 8-weeks post-implantation, and regenerated muscle tissues were harvested for histological and immunofluorescence staining. Prior to tissue collection, electrophysiological and force measurements were conducted. Under anesthesia, the experimental muscle was exposed. Stimulating needle electrodes were inserted at both ends of the regenerated muscle, and compound muscle action potentials (CMAPs) were recorded using a physiological signal acquisition system (RM-6240, Chengdu Instrument Factory, Chengdu, China). Following euthanasia, muscle tissues were harvested, and tetanic force was recorded using a mechanical testing apparatus under electrical stimulation (50 V) applied via the electrophysiology setup. Histological staining Harvested muscle tissues were fixed in 4% paraformaldehyde for 24 h, followed by cryosectioning. H&E and Masson's trichrome staining were performed according to the manufacturer's protocols. Images were captured using a Leica microscope (DM3000, Germany). Immunofluorescence staining Muscle sections were rehydrated and blocked with 5% goat serum, then incubated overnight at 4 °C with the following primary antibodies: NF09 mouse monoclonal (1:200, Abcam, ab7794), Pax7 rabbit polyclonal (1:100, Cohesion, CQA3315), Desmin rabbit polyclonal (1:50, Abcam, ab15200), α-SMA rabbit polyclonal (1:300, Abcam, ab7817), CD206 rabbit polyclonal (1:500, Abcam, ab64693), and iNOS rabbit polyclonal (1:200, Abcam, ab15323). After PBS washes, sections were incubated with corresponding secondary antibodies for 2 h, washed again, and mounted with DAPI. Fluorescence images were acquired using a confocal laser scanning microscope (Leica SP8, Germany). Single-cell sequencing analysis mFS and mDS were implanted into the tibialis anterior muscle of a rat VML model, with samples harvested after two weeks. The retrieved tissue was subsequently homogenized for nuclei extraction. Single Nuclei RNA libraries were prepared using SeekOne® DD Single Cell 3’ library preparation kit (SeekGene Catalog No.K00202). The libraries were then sequenced on illumina NovaSeq X Plus with PE150 read length. Processing, analysis, and visualization of the raw data were performed using the SeekSoul online analysis platform (SeekSoul, China). Immunofluorescence staining was performed on rat VML model sections two weeks post-injury using a rabbit polyclonal anti-Hif1a primary antibody (1:100, Cohesion, CPA3239). Detailed primer sequences of Pax7, Foxp3, and Top2a used for RT-qPCR in the VML model at 2 weeks for mechanism validation are available in Supplementary Table 1. Statistical analysis All quantitative results were obtained from at least three independent samples. Data were presented as the mean ± s.e.m. Statistical analyses were performed using GraphPad Prism 7 (GraphPad Software, USA). Unpaired Student's t-test was used for single comparisons, one-way ANOVA with Tukey's post hoc test for multiple comparisons, and two-way ANOVA with Tukey's post hoc test for multivariable comparisons. Quantification of samples was carried out using ImageJ software v1.5. For all experiments, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 were considered statistically significant; ns indicates not significant. CRediT authorship contribution statement Guangzhou Song: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Wenqian Cong: Methodology, Investigation. Hongbai Lu: Methodology, Investigation. Jiatong Wang: Methodology, Investigation. Yumeng Wang: Methodology, Investigation. Yanzhen Zhao: Methodology, Investigation. Shaowen Wang: Methodology, Investigation. Shijie Zhu: Methodology, Investigation. Adam C. Midgley: Writing – review & editing, Writing – original draft. Wen Li: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Meifeng Zhu: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Ethics approval and consent to participate All animal procedures were approved by the Animal Ethics Committee of Nankai University, Tianjin (Approval No. 22025-SYDWLL-000495) and conducted in accordance with the Guide for the Care and Use of Laboratory Animals. A total of 42 male Sprague Dawley rats aged 8-10 weeks (weight range: 280-320 g) were obtained from SPF Biotechnology Co., Ltd. (Beijing, China). Rats were randomly housed in cages (three animals per cage) with ambient temperature maintained at 23 ± 3 °C, relative humidity at 40-70%, and a 12-h light/dark cycle. Standard food and water were provided ad libitum. Before surgery, rats were fasted for 12 h with free access to water. Postoperatively, all animals were placed on heating pads to maintain stable respiration and body temperature. Twelve rats were used for subcutaneous implantation experiments, and 30 rats were used for the VML model. In all experiments, data were obtained with a sample size of n = 3 per group. Funding This work was financially supported by the National Key Research and Development Program of China (2025YFF1505200), National Natural Science Foundation of China (82272156), Tianjin Natural Science Foundation (24JCQNJC01520), Haihe Laboratory of Sustainable Chemical Transformations(25HHWCSS00020), and Tianjin Science and Technology Program (Grant No.23JCZDJC00810). 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. One sentence summary: Fabrication of decellularized skeletal muscle extracellular matrix into tissue fabric scaffolds for promotion of muscle defect repair. Footnotes Peer review under the responsibility of editorial board of Bioactive Materials. Supplementary data to this article can be found online at https://doi.org/10.1016/j.bioactmat.2026.05.002. Contributor Information Wen Li, Email: liwen2410@tju.edu.cn. Meifeng Zhu, Email: zhumeifeng@nankai.edu.cn. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 References Associated Data Supplementary Materials Multimedia component 1