A Muscle‐Mimetic Core‐Sheath Composite Yarn Scaffold for In‐Body Tissue Induction and Regeneration of Small‐Diameter Vascular Grafts

ABSTRACT

Cardiovascular diseases remain a leading cause of mortality worldwide. Small‐diameter vascular grafts (SDVGs) continue to face critical clinical challenges, including acute thrombosis, intimal hyperplasia, and insufficient endothelialization. Inspired by the hierarchical structure of skeletal muscle, in which myofibrils assemble into muscle fibers and then into functional tissue, we developed a novel tissue‐engineered vascular graft (TEVG) based on a muscle‐mimetic composite yarn. Using friction spinning technology, we fabricated a core‐sheath composite yarn with a poly(ethylene terephthalate) (PET) filament as the artificial myofibril core to provide durable mechanical support, and poly(glycolic acid) (PGA) staple fibers as the sheath component to replicate the extracellular matrix (ECM) topology of muscle fibers, thereby enhancing bioactivity. The knitted tubular scaffold was implanted subcutaneously in rabbits for in vivo tissue induction, followed by decellularization, yielding an extracellular matrix‐rich and biocompatible graft. This “yarn mimicking muscle, fabric transforming into vessel” strategy achieved staged vascular regeneration. In a canine carotid artery replacement model, the TEVG maintained 100% patency at one month, with histological evidence of endothelialization (CD31+), smooth muscle regeneration (α‐SMA+), and collagen deposition. This bioinspired approach demonstrates short‐term feasibility for small‐diameter vascular regeneration in a canine model, providing a promising platform for long‐term studies and clinical translation.

A muscle‐mimetic composite yarn is fabricated using friction spinning, with a PET filament core providing mechanical support and a PGA staple fiber sheath enhancing bioactivity. After subcutaneous implantation and decellularization, an extracellular matrix‐rich graft is obtained. In a canine carotid artery model, the tissue‐engineered vascular graft shows good patency, with evidence of endothelialization, smooth muscle regeneration, and collagen deposition.

1. Introduction

Cardiovascular diseases remain the leading cause of mortality worldwide, and the management of end‐stage vascular disease relies heavily on surgical revascularization using vascular grafts [1, 2, 3]. However, the clinical application of small‐diameter (< 6 mm) vascular grafts still faces significant challenges: autologous vessels are limited in supply and involve secondary trauma, while existing synthetic grafts (such as ePTFE or PET) are prone to failure under small‐diameter conditions due to thrombosis and intimal hyperplasia [4, 5, 6]. Tissue engineering offers a promising alternative strategy by constructing biomimetic scaffolds that recapitulate the architecture of the native extracellular matrix (ECM), thereby providing a new pathway for vascular regeneration [7, 8, 9, 10]. These scaffolds can guide the directed migration, proliferation, and differentiation of host cells, enabling the in situ regeneration of functional tissue and overcoming limitations related to donor scarcity and immune rejection [11, 12, 13]. With the increasing burden of cardiovascular diseases and the broad prospects of tissue engineering in organ reconstruction, the development of small‐diameter tissue‐engineered blood vessels (TEBVs) that combine favorable mechanical properties with bioactivity has become a research hotspot in regenerative medicine.

Among these emerging strategies in vascular tissue engineering, in situ tissue induction has shown considerable promise for clinical translation [14, 15, 16, 17, 18]. This technique involves implanting a biodegradable scaffold into a host (e.g., subcutaneously), leveraging the host's innate repair mechanisms to guide cell infiltration and ECM deposition within the scaffold, thereby forming a living biological tube. Subsequent decellularization yields an ECM‐based scaffold that retains native bioactive components. This method not only avoids the complexity and variability associated with in vitro cell culture but also recapitulates key biological and structural characteristics of native vessels, promoting rapid endothelialization and functional remodeling upon implantation [19, 20, 21]. Zhi et al. [22], inspired by steel‐reinforced concrete in architectural engineering, developed mechanically reinforced biotubes (hMPBs) by subcutaneous implantation of a melt‐spun poly(ε‐caprolactone) fiber skeleton. These hMPBs demonstrated superior mechanical strength, puncture resistance, vascular regeneration, and long‐term patency in rat, canine, and sheep models, showing great promise as clinical alternatives for arterial replacement and arteriovenous grafting.

The success of in situ tissue induction strategies critically depends on the rational design of the initial scaffold. The scaffold must provide sufficient mechanical support in the early post‐implantation phase to maintain lumen patency, while also offering a favorable biointerface to guide cell behavior and facilitate ECM deposition [23, 24]. Currently, single‐material systems struggle to fulfill both requirements simultaneously. For instance, poly(glycolic acid) (PGA) exhibits good cell affinity and controllable degradation, but suffers from poor mechanical strength and rapid degradation rates [25, 26, 27, 28, 29]. In contrast, poly(ethylene terephthalate) (PET) demonstrates excellent mechanical durability, yet its surface is relatively bioinert, hindering cell adhesion and tissue integration [30].

From the perspective of textile materials science, the micro‐topological structure of fibers is a critical determinant of cell behavior. Filaments have smooth surfaces and low specific surface areas, offering limited cell anchoring sites. In contrast, staple fibers, through stacking and entanglement, form three‐dimensional porous structures that significantly increase specific surface area and surface roughness, more closely resembling the fibrous network of the native extracellular matrix (ECM). This “staple fiber effect” not only provides more anchoring sites for cell adhesion but also promotes nutrient transport and cell migration, thereby enhancing the overall bioactivity of the scaffold.

Notably, the hierarchical architecture of skeletal muscle offers a novel bioinspired paradigm for scaffold design [31]. Native skeletal muscle features a precise, multiscale hierarchical assembly from the micro to macroscale: contractile protein‐based myofibrils are ensheathed by connective tissue to form muscle fibers, which further aggregate into fascicles, ultimately integrating into a functional whole muscle with contractile capacity (Figure 1a). This cross‐scale “basic unit‐intermediate fiber‐macroscopic bundle–functional tissue” structure is critical for mechanical transmission, cellular support, and tissue regeneration. Inspired by this, we propose a “yarn‐mimicking muscle” design concept: PET filaments, which provide long‐term mechanical support, act as the “artificial myofibril” skeleton, while PGA staple fibers with superior cell affinity are helically integrated around the PET core to form a core‐sheath composite yarn with a biomimetic muscle fiber‐like architecture. The PGA staple fiber layer not only mimics the topological structure of native muscle ECM, providing abundant sites for cell adhesion and migration, but also guides the ingrowth and remodeling of neotissue in a spatiotemporal manner via its gradient degradation behavior. Subsequently, through knitting, subcutaneous implantation‐induced in situ tissue regeneration, and decellularization, this bioinspired muscle‐mimicking yarn is ultimately transformed into a decellularized biotube (DBT) with a functional vascular wall structure, achieving the regenerative goal of “fabric‐to‐vessel transformation”.

FIGURE 1.

Schematic diagram of the fabrication process for the tissue‐engineered vascular graft.

Although previous studies have explored composites of degradable and non‐degradable materials for vascular constructs, systematic investigation into leveraging the topological properties of PGA staple fibers to mimic the hierarchical structure of muscle and thereby regulate cellular responses for TEBV fabrication remains limited. To address this issue, we propose a novel in situ‐induced tissue‐engineered vascular graft based on a PET/PGA core‐sheath composite yarn (Figure 1b). Using friction spinning technology, a PET filament serves as the core to provide durable mechanical support, while PGA staple fibers form the sheath to create a bioactive interface. This design leverages the topological advantages of PGA staple fibers to activate host cell responses and utilizes their staged degradation behavior in coordination with the PET skeleton to achieve sequential vascular regeneration. The knitted tubular scaffold is implanted subcutaneously in rabbits; degradation of the PGA sheath recruits host cells and induces ECM deposition, forming a bioactive interface on the PET framework. After decellularization, an ECM‐rich graft is obtained for vascular replacement. Upon implantation, residual PGA continues to degrade, creating space for cell infiltration, while the ECM‐coated PET fibers act as stable anchors to support tissue remodeling.

This study systematically evaluates the physicochemical and biological properties of the materials from the yarn stage to the final graft, and validates graft patency and tissue regeneration capacity in a large animal model. By integrating principles of textile structural design, composite material synergy, and regenerative medicine, this strategy offers a novel and clinically promising approach for small‐diameter vascular regeneration.

2. Materials and Methods

2.1. Materials

Polyglycolide (PGA) pellets were purchased from Inner Mongolia Pujing Polymer Materials Technology Co., Ltd. (China). Polyethylene terephthalate (PET) multifilament yarns (138 dtex/36 f) were supplied by Jiangsu Shenghong Fiber Co., Ltd. (China). All chemicals and reagents for biological experiments, including phosphate‐buffered saline (PBS), Dulbecco's Modified Eagle Medium (DMEM), fetal bovine serum (FBS), penicillin/streptomycin (PS), Calcein‐AM/PI live/dead cell staining kit, Cell Counting Kit‐8 (CCK‐8), Triton X‐100, and 4% paraformaldehyde (PFA) were obtained from Thermo Fisher Scientific (USA) and Biosharp (China).

Murine fibroblast L929 cells (RRID:CVCL_0462, catalog No. SNL‐402) and human umbilical vein endothelial cells HUVEC (RRID:CVCL_2959, catalog No. SNP‐H204) were purchased from Wuhan Shangen Biotechnology Co., Ltd., China. Antibodies against CD31 and α‐SMA were obtained from Wuhan Kaishengxiang Biotechnology Co., Ltd., China.

All reagents were of analytical or biological grade and used as received.

2.2. Preparation of PGA Spunbond Fabric and Composite Yarns

Inspired by the hierarchical architecture of skeletal muscle, we designed a core‐sheath composite yarn to replicate the structural organization of muscle fibers. PGA spunbond fabric was manufactured using a commercial spunbond machine (F‐6C, Yantai Huada Technology Co., Ltd., China). The detailed process included melt extrusion, spinning, drawing, web laying, and thermal bonding. The specific process parameters are summarized in Table 1.

TABLE 1.

Spunbond experiment parameters.

Spinning assembly	Transfer line	Barrel zone 1	Barrel zone 2	Barrel zone 3	Barrel zone 4	Quench air	Smooth roll	Embossed roll
235°C	235°C	235°C	235°C	235°C	230°C	20°C	58°C	54°C

The obtained PGA spunbond fabric was cut into short fibers (25–35 mm in length). These PGA short fibers were then opened and carded using a miniature opening machine (HFX‐A0, Suzhou Huafei Textile Technology Co., Ltd., China) and a carding machine (HFX‐A1, Suzhou Huafei Textile Technology Co., Ltd., China) to form PGA slivers.

Core‐sheath structured PET/PGA composite yarns were produced on a friction spinning machine (HFX‐02, Suzhou Huafei Textile Technology Co., Ltd., China). The PET multifilament was used as the core component (analogous to the artificial myofibril skeleton), while the PGA sliver was fed as the sheath component (mimicking the ECM topology around muscle fibers). By varying the feeding speed of the PGA sliver to 0.2, 0.3, and 0.4 m/min, three types of composite yarns with different PGA contents were fabricated and designated as CY‐0.2, CY‐0.3, and CY‐0.4. Other spinning parameters were set as follows: cylinder speed of 3500 rpm, friction roller speed of 4000 rpm, and output speed of 7 m/min.

2.3. Fabrication and Characterization of Tubular Fabrics

The composite yarns were knitted into tubular scaffolds using a hand‐operated flat‐bed knitting machine (12‐gauge). This knitting process transformed the muscle‐mimetic yarns into a three‐dimensional tubular scaffold, serving as the precursor for subsequent in vivo tissue induction. The knitted tubes were then placed over a 6 mm diameter stainless‐steel mandrel, overwrapped with aramid yarn in a spiral pattern, and heat‐set in an oven at 180°C for 5 min to obtain the final tubular fabrics. The tubular fabrics woven from composite yarns CY‐0.2, CY‐0.3, and CY‐0.4 were respectively named as TF‐0.2, TF‐0.3, and TF‐0.4.

The surface morphology of the original PGA spunbond fabric and the composite yarns was observed using a scanning electron microscope (SEM, JSM‐5600LV, JEOL, Japan) and a 3D digital microscope (HIROX RH2000, HIROX Co., Ltd., Japan). The linear density of the yarns was measured by the cut‐and‐weigh method. The mechanical properties (breaking strength and elongation) of the yarns and tubular fabrics were tested on an electronic single‐yarn strength tester (YG061F, Laizhou Electronic Instrument Co., Ltd., China) and a universal testing machine (EUT2203, Shenzhen Sansi Testing Technology Co., Ltd., China), respectively. The hairiness and evenness of the yarns were evaluated using a fully automatic hairiness tester (CFH) and an evenness tester (CFE500), respectively (both from Suzhou Changfeng Textile Electromechanical Technology Co., Ltd., China). Thermogravimetric analysis (TGA) was performed on a TGA5500 instrument (TA Instruments, USA) under a nitrogen atmosphere from 30°C to 600°C at a heating rate of 10°C/min.

2.4. In Vitro Degradation and Cytocompatibility Evaluation

The in vitro degradation behavior of the tubular fabrics was investigated by incubating samples in PBS (pH 7.4) at 37°C for up to 35 days. The mass loss was calculated at predetermined time intervals (5, 15, 25, 35 days) according to Equation (1). The morphological changes of the samples during degradation were observed under a 3D microscope.

$$M\left(\%\right)=\frac{M_0-M_1}{M_0}\times 100\%$$ (1)

where M is the mass loss rate (%), M₀ is the initial weight of the sample (g), and M₁ is the mass after degradation (g).

Cytocompatibility was assessed using murine fibroblast L929 cells. The tubular fabrics were sterilized with 75% ethanol, rinsed with PBS, and then immersed in culture medium to prepare extracts (0.2 g/mL). L929 cells were cultured in the extracts. Cell viability was quantified using the CCK‐8 assay after 1, 3, and 5 days of culture. Cell activity was calculated according to Equation (2). Furthermore, live/dead staining and cytoskeleton/nucleus staining (using FITC‐phalloidin and DAPI) were performed and visualized under a laser scanning confocal microscope (AX, Nikon Instruments (Shanghai) Co., Ltd., China) to observe cell morphology and distribution.

$$C\mathrm{e}ll\mathrm{viability}\left(\%\right)=\frac{{\mathrm{A}}_{\mathrm{sample}}-{\mathrm{A}}_{\mathrm{blank}}}{{\mathrm{A}}_{\mathrm{control}}-{\mathrm{A}}_{\mathrm{blank}}}\times 100\%$$ (2)

where A_sample is the absorbance value of the sample group, A_blank is the absorbance value of the blank group, and A_control is the absorbance value of the control group.

2.5. Subcutaneous Implantation, Decellularization, and Evaluation of Decellularized Biotubes

A composite scaffold, consisting of a 3D‐printed polylactic acid (PLA) inner tube and an outer stainless‐steel spring, was designed to provide mechanical support during implantation. The tubular scaffolds were placed over this composite mandrel and surgically implanted into subcutaneous dorsal pockets in New Zealand White rabbits. Male New Zealand White rabbits (n = 6, 5 months old, 3.0–3.5 kg) were purchased from Hubei Yizhicheng Biotechnology Co., Ltd. The animals were kept under standard laboratory conditions with unlimited access to food and water. Randomization was applied for group allocation. Humane endpoints included > 20% body weight loss or moribund state, at which point animals were euthanized. After 4 weeks, the constructs (named as biotubes, BT) were explanted.

The explanted biotubes were decellularized using a multi‐step protocol involving agitation in solutions containing CHAPS, thiophene tridecyl ester, ASB‐14, SB3‐10, and Benzonase nuclease to remove cellular components while preserving the extracellular matrix (ECM) [32]. The resulting acellular grafts were termed decellularized biotubes (DBT).

The morphology and surface structure of the DBTs were characterized using optical microscopy and SEM. Their radial mechanical properties were tested using a universal testing machine. The efficiency of decellularization and the retention of collagen were verified by hematoxylin and eosin (H&E) staining and Masson's trichrome staining of histological sections.

The cytocompatibility of the DBTs was further evaluated using human umbilical vein endothelial cells (HUVECs) via CCK‐8 assay and live/dead staining, as described in Section 2.4.

2.6. In Vitro Blood Compatibility Assessment

The anticoagulant properties of the DBTs were evaluated by measuring key coagulation parameters. Platelet‐poor plasma (PPP) was isolated from rabbit whole blood by centrifugation. The DBT samples were incubated with PPP at 37°C for 30 min. After incubation, the activated partial thromboplastin time (APTT), prothrombin time (PT), thrombin time (TT), and fibrinogen (FIB) content of the plasma were measured using a fully automated coagulation analyzer (CS‐5100, Sysmex Corporation, Japan).

2.7. In Vivo Evaluation in a Canine Carotid Artery Model

The in vivo performance of the best‐performing graft (DBT‐0.2) was evaluated in a canine carotid artery replacement model. Healthy male beagle dogs (n = 3, 14 months old, 13 kg) were obtained from Hunan Hengxinwang Biotechnology Co., Ltd.

The animals were kept under standard laboratory conditions with unlimited access to food and water. Humane endpoints included severe distress, respiratory distress, or inability to stand, at which point animals were humanely euthanized. A segment of the right/left carotid artery was surgically exposed, clamped, and resected. A DBT‐0.2 graft of matching diameter was anastomosed end‐to‐end to the native artery using continuous sutures. Patency was confirmed immediately after surgery.

After one month, graft patency and blood flow were assessed using color Doppler ultrasound (DC‐N35, Mindray Medical International Ltd., China). The animals were then euthanized, and the explanted grafts were harvested for histological analysis. The grafts were fixed, paraffin‐embedded, sectioned, and stained with H&E and Masson's trichrome to observe tissue structure and collagen deposition. Immunofluorescence staining for CD31 (an endothelial cell marker) and α‐SMA (a smooth muscle cell marker) was performed to evaluate endothelialization and smooth muscle regeneration within the graft walls.

2.8. Ethical Certification

All animal experiments in this study were conducted in accordance with the guidelines and approved by the Wuhan Institutional Animal Care and Use Committee of Huazhong University of Science and Technology, China (IACUC approval number: 4527).

2.9. Statistical Analysis

All quantitative data are presented as mean ± standard deviation (SD) from at least three independent experiments (n ≥ 3), unless otherwise stated in the figure legends. No outlier exclusion or data transformation was applied. For comparisons between two groups, an unpaired two‐tailed Student's t‐test was used. For three or more groups, statistical significance was assessed using one‐way analysis of variance (ANOVA) followed by Tukey's post hoc test. Differences were considered significant at p < 0.05. Significance levels are indicated in the figures as ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001. The schematic illustrations and ToC figure were created using Adobe Photoshop 2024 (25.5) and Cinema 4D 2024 (2024.3). The graphs and statistical analysis were analyzed in GraphPad Prism software (9.0.0.121) and Origin 2018 (9.50.00).

3. Results and Discussion

3.1. Preparation and Properties of PGA Staple Fiber‐Reinforced Composite Yarns

A primary challenge in constructing high‐performance TEBVs lies in developing scaffold materials that simultaneously combine excellent mechanical properties with intrinsic bioactivity. PET possesses outstanding mechanical strength and long‐term stability, but its bioinertia often leads to thrombosis and poor tissue integration. In contrast, PGA exhibits excellent biocompatibility and controllable degradation characteristics, which promote cell adhesion and tissue regeneration. However, its inherent brittleness and excessively rapid degradation rate limit its standalone application as a load‐bearing vascular scaffold. In this study, we employed friction spinning technology to integrate PGA staple fibers as a sheath onto a PET core yarn, aiming to synergistically leverage the mechanical advantages of PET and the biological functions of PGA. This core–sheath architecture directly mimics the hierarchical organization of skeletal muscle, where the PET core acts as an “artificial myofibril” skeleton, and the PGA sheath replicates the ECM topology around muscle fibers.

We first prepared PGA spunbond nonwoven fabric via the spunbond process (Figure 2a). Figure S1 illustrates the preparation procedure of the spunbond method. Scanning electron microscopy (SEM) revealed that the PGA spunbond fabric consists of randomly interwoven PGA fibers forming a porous three‐dimensional network structure (Figure 2b). The PGA fibers exhibited a smooth surface, and their diameter distribution followed a normal distribution (Figure 2c), with an average diameter of 19.16 ± 3.15 µm, indicating a stable spinning process and uniform fiber morphology. By adjusting the feeding speed of the PGA sliver during friction spinning, we successfully fabricated three types of composite yarns with different PGA contents (Figure 2d). As the PGA feeding speed increased, the linear density of the composite yarns significantly rose (Figure 2f), from 93.4 tex for CY‐0.2 to 230 tex for CY‐0.4. The mass ratio of PGA to PET showed a gradient variation (Figure S2), successfully achieving precise control over the material composition. The cross‐sectional structure (Figure 2e, top) and surface morphology (Figure 2e, bottom) of the composite yarns clearly demonstrated the successful coating of the PET core yarn by PGA staple fibers, forming a muscle fiber‐like morphology that provides both mechanical integrity and bioactive interfaces. With increasing PGA content, the yarn surface became fuller, and the distribution of PGA staple fibers became more uniform and dense.

FIGURE 2.

Morphology, structure, and basic properties of PGA spunbond nonwoven fabric and composite yarns. (a) Optical image of PGA spunbond nonwoven fabric; (b) Scanning electron microscopy (SEM) image of PGA spunbond nonwoven fabric; (c) Histogram of PGA fiber diameter distribution (n = 100 fibers); (d) Optical image of composite yarns; (e) SEM image (top) and 3D microscopy image (bottom) of composite yarns; (f) Linear density of composite yarns (n = 5); (g) Breaking strength and elongation at break of composite yarns (n = 5); (h) Hairiness index of composite yarns (n = 5); (i) Yarn evenness of composite yarns (n = 5); (j) Thermogravimetric analysis (TGA) of composite yarns (n = 5). Data are presented as mean ± SD. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

The mechanical testing results (Figure 2g) revealed the reinforcing effect of PGA staple fibers. The pure PET yarn exhibited a breaking strength of 24.15 ± 0.47 cN/tex and an elongation at break of 20.43 ± 0.81%. In comparison, all composite yarns showed significant improvements in both breaking strength and elongation at break (CY‐0.2: 29.66 ± 0.32 cN/tex, 24.43 ± 1.01%; CY‐0.3: 30.91 ± 0.52 cN/tex, 24.73 ± 0.77%; CY‐0.4: 32.76 ± 1.00 cN/tex, 26.31 ± 0.90%). This enhancement can be primarily attributed to the increased inter‐fiber friction and cohesion after the incorporation of PGA staple fibers, which allows stress to be transferred and dispersed more effectively between the PET core and the PGA sheath under external load. This core‐sheath design thus achieves a synergistic combination of mechanical robustness (from PET) and structural flexibility (from PGA), which is difficult to realize with either material alone. However, the increase in PGA content also led to a higher degree of yarn hairiness (Figure 2h), which may affect the surface smoothness of subsequent fabrics. Notably, the yarn evenness improved with increasing PGA content (Figure 2i), as indicated by the decrease in both the unevenness (U%) and coefficient of variation (CV%), suggesting that higher PGA content contributes to a more uniform yarn structure. Thermogravimetric analysis (TGA) demonstrated the thermal stability of the materials (Figure 2j). The thermal decomposition behavior of all composite yarns fell between that of pure PGA (onset decomposition temperature ∼ 300°C) and pure PET (onset decomposition temperature ∼ 370°C). With increasing PGA content, the initial weight loss temperature slightly decreased, which is consistent with the higher proportion of the PGA component in the composites.

3.2. Fabrication, Properties, and In Vitro Biocompatibility of Tubular Fabrics

Having characterized the physicochemical properties of the composite yarns, we next transformed them into three‐dimensional tubular scaffolds and evaluated their structural and biological performance. Converting yarns into tubular scaffold structures represents a critical step toward replicating the tubular morphology of native blood vessels. Using weft knitting technology, the composite yarns were knitted into tubular fabrics (Figure S3). After scouring and heat‐setting treatments, the knitted tubes formed stable tubular fabrics with an inner diameter of 6 mm (Figure 3a,b). All tubular fabrics exhibited well‐defined knitted structures and uniform tubular morphology (Figure S4). This knitting process transformed the muscle‐mimetic yarns into a three‐dimensional scaffold, representing the transition from “yarn mimicking muscle” to “fabric precursor for vascular regeneration”.

FIGURE 3.

Evaluation of degradation behavior, cyclic mechanical properties, and in vitro cytocompatibility of tubular fabrics. (a) Preparation of tubular fabric before heat setting; (b) Optical image of tubular fabric; (c) Weight loss rate of tubular fabrics after degradation in PBS solution for 5, 15, 25, and 35 days (n = 5); Axial tensile mechanical properties of tubular fabrics: (d) Stress–strain curves, (e) Breaking strength and elongation at break (n = 5), (f) Stress–strain curves after 50 cycles of 30% axial tensile strain, (g) Stress–strain curves after 50 cycles of 50% axial tensile strain; Radial tensile mechanical properties of tubular fabrics: (h) Stress–strain curves, (i) Breaking strength and elongation at break (n = 5), (j) Stress–strain curves after 50 cycles of 30% radial tensile strain, (k) Stress–strain curves after 50 cycles of 50% radial tensile strain; (l) Fluorescence imaging of L929 cell proliferation; (m) Viability of L929 cells cultured with extracts of different tubular fabrics (n = 6). Data are presented as mean ± SD. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

To evaluate their potential as degradable scaffolds, in vitro degradation tests were conducted on the tubular fabrics. After 35 days of immersion in PBS solution, all samples showed significant mass loss (Figure 3c). More importantly, the degradation rate was positively correlated with the PGA content. The rapid mass loss within the first 15 days corresponded to the hydrolysis of the amorphous regions of PGA, while the slower mass loss between days 15 and 35 was attributed to the gradual exposure of the PET framework (Figure S5). TF‐0.4 exhibited the highest mass loss rate, whereas TF‐0.2 showed the lowest. These results clearly demonstrate that the degradation kinetics of the scaffold material can be precisely controlled by adjusting the content of PGA staple fibers. This tunable degradation behavior is crucial, as it ideally requires the scaffold to provide sufficient mechanical support initially and then gradually degrade to make room for host tissue integration and regeneration. The faster degradation of TF‐0.4, while leading to higher mass loss, also resulted in compromised structural integrity after 35 days, suggesting that an optimal PGA content (i.e., TF‐0.2) is necessary to balance degradation rate with long‐term scaffold stability.

We further systematically evaluated the mechanical properties of the tubular fabrics (Figure 3d–k). Both axial and radial tensile tests revealed significant anisotropy. Under axial tension, TF‐0.4 exhibited the highest breaking strength (Figure 3e), consistent with the superior strength of the CY‐0.4 yarn itself. Notably, all fabrics maintained good structural integrity and fatigue resistance after 50 cycles of tensile loading at 30% and 50% strain (Figure 3f,g,j,k), with minimal changes in hysteresis loops, indicating their ability to withstand cyclic mechanical stresses induced by pulsatile blood flow.

In vitro biocompatibility is a gold standard for evaluating biomaterials. We assessed the cytotoxicity of the tubular fabric extracts using L929 fibroblast cells. Live/dead cell staining (Figure 3l) revealed a majority of green‐stained (live) cells and only a minimal number of red‐stained (dead) cells in the field of view. The CCK‐8 assay results (Figure 3m) further supported this, showing that the cell viability after 1, 3, and 5 days of culture in all three extract groups exceeded 80%, meeting the requirements for non‐cytotoxicity according to the international standard (ISO 10993–5). Although a slightly lower cell proliferation activity was observed in the TF‐0.4 group on day 3—potentially due to the higher concentration of initial degradation products—by day 5, all groups exhibited robust cell growth, indicating favorable cytocompatibility of the materials.

3.3. In Vivo Tissue Induction, Decellularization, and Functional Evaluation of Decellularized Biotubes

With their favorable in vitro performance confirmed, these tubular scaffolds were subjected to in vivo tissue induction to generate ECM‐rich decellularized biotubes. A core challenge in small‐diameter tissue‐engineered blood vessels (TEBVs) is effectively preventing acute thrombosis and promoting rapid endothelialization. Our strategy leverages the host's in vivo environment as a “bioreactor” to generate natural extracellular matrix (ECM) within the scaffold, followed by decellularization to produce an ECM‐rich, low‐immunogenicity graft.

3.3.1. Preparation and Properties of Biotubes and Decellularized Biotubes

The tubular fabrics were sheathed onto specially designed composite PLA mandrels and implanted subcutaneously in rabbits for a 4‐week in vivo induction period (Figure 4a,b). During this time, host cells infiltrated the fabrics and secreted substantial ECM components. The resulting biotubes were then decellularized using a combination of chemical agents to obtain acellular biotubes (Figure 4c).

FIGURE 4.

Morphology and radial mechanical properties of decellularized biotubes. (a) Sample preparation before subcutaneous implantation; (b) Sutured incision after subcutaneous implantation; (c) Optical images of decellularized biotubes; (d) Scanning electron microscopy (SEM) images of the inner and outer surfaces of decellularized biotubes; Radial tensile mechanical properties of decellularized biotubes: (e) Stress–strain curves, (f) Breaking strength and elongation at break (n = 5), (g) Stress–strain curves after 50 cycles of 50% radial tensile strain. Data are presented as mean ± SD. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

After decellularization, variations in structural integrity were observed among the decellularized biotubes (Figure 4c). DBT‐0.2 retained the most intact tubular morphology, demonstrating that a balanced muscle‐mimetic design (sufficient PET core content with optimal PGA sheath degradation) is critical for preserving the hierarchical scaffold architecture after decellularization. In contrast, the higher PGA content in DBT‐0.3 and DBT‐0.4 led to excessive degradation during the 4‐week subcutaneous implantation, resulting in a weakened PET framework that could not adequately support the ECM after removal of cellular components. This explains the observed decrease in radial breaking strength for DBT‐0.3 and DBT‐0.4 compared to DBT‐0.2. SEM images (Figure 4d) revealed a continuous ECM layer covering both the inner and outer surfaces of all decellularized biotubes. This outcome can be explained by the degradation behavior of the materials: after 4 weeks of implantation, TF‐0.4 with high PGA content underwent more substantial degradation, compromising the integrity of the PET framework. As a result, its structure could not be maintained after cell removal. In contrast, DBT‐0.2, with the highest PET content, retained the strongest mechanical skeleton, thereby providing better structural support to the ECM after decellularization. These observations were corroborated by radial tensile tests (Figure 4e,f), which showed that DBT‐0.2 had significantly higher radial breaking strength than DBT‐0.3 and DBT‐0.4. All DBTs exhibited exceptionally high elongation at break (>220%) and maintained stable mechanical properties after 50 cycles of tensile loading (Figure 4g), demonstrating compliance and fatigue resistance comparable to those of native blood vessels.

3.3.2. ECM Composition, In Vitro Biocompatibility, and Anticoagulant Properties of Decellularized Biotubes

Histological staining (H&E and Masson's trichrome) clearly illustrated the process of in vivo induction and subsequent decellularization (Figure 5a). Biotubes showed numerous cell nuclei (blue) and collagen fibers (blue), while after decellularization, cell nuclei were completely removed, and ECM components such as collagen fibers were well preserved. This confirms the high efficiency and success of our decellularization protocol.

FIGURE 5.

Section staining, in vitro cytocompatibility, and anticoagulant properties of decellularized biotubes. (a) H&E staining and Masson's trichrome staining of biotubes and decellularized biotubes; (b) Fluorescence imaging of HUVEC proliferation; (c) Viability of HUVECs cultured with extracts of different decellularized biotubes (n = 6); In vitro coagulation test results of decellularized biotubes: (d) Activated partial thromboplastin time (APTT), (e) Prothrombin time (PT), (f) Thrombin time (TT), (g) Fibrinogen (FIB) concentration (n = 3). Data are presented as mean ± SD. ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001.

In vitro coagulation assays revealed that all DBT samples, after incubation with platelet‐poor plasma (PPP), significantly prolonged the activated partial thromboplastin time (APTT), prothrombin time (PT), and thrombin time (TT) compared to the blank control (Figure 5d–f). This suggests effective inhibition of both the intrinsic and extrinsic coagulation pathways. Additionally, a significant decrease in fibrinogen (FIB) concentration was observed (Figure 5g), indicating interference with the final stage of the coagulation cascade. These results demonstrate outstanding anticoagulant properties of the DBTs, likely attributable to the preserved natural ECM components and unique microstructural features on their surface, which collectively suppress platelet adhesion and activation of the coagulation cascade. This functionality is essential for small‐diameter TEBVs to prevent acute post‐implantation thrombosis.

The in vitro biocompatibility of the decellularized biotubes (DBTs) was further evaluated using human umbilical vein endothelial cells (HUVECs). Live/dead staining and CCK‐8 assay results (Figure 5b,c) demonstrated that extracts from the DBTs well supported HUVEC proliferation and viability, with robust cell growth over 5 days, indicating excellent cytocompatibility.

Due to its optimal mechanical properties and structural integrity, DBT‐0.2 was selected as the representative group for subsequent in vivo experiments. It not only retained sufficient mechanical strength to withstand blood pressure but also exhibited superior anticoagulant capacity and cell affinity conferred by the surface ECM coating, thereby effectively addressing the dual challenges of synthetic material bio‐inertness and the poor mechanical performance of natural materials. This outcome validates our muscle‐inspired strategy: the PET core provides structural “skeletal” strength, while the PGA‐induced ECM coating confers “functional” bioactivity, together mirroring the structure‐function duality inherent in native muscle tissue.

3.4. In Vivo Biocompatibility of the Tissue‐Engineered Vascular Graft

To evaluate the performance of our designed tissue‐engineered vascular graft (DBT‐0.2) in a realistic physiological environment, we conducted a one‐month carotid artery replacement experiment in beagle dogs (Figure S6). All animals survived the surgical procedure without complications, including acute thrombosis, anastomotic aneurysm formation, or life‐threatening hemorrhage, demonstrating favorable surgical handling and short‐term biosafety of the graft.

Graft patency and hemodynamic performance were the primary endpoints. Color Doppler ultrasound examination at one month post‐implantation (Figure 6a,b) showed that the implanted DBT‐0.2 segment remained fully patent. The graft demonstrated continuous, uniform color flow signals comparable to those observed in the native carotid artery (Figure 6b), with no significant turbulence or retrograde flow detected, indicating that the graft did not induce abnormal flow patterns that could lead to thrombosis. Spectral Doppler analysis revealed a typical arterial triphasic waveform with a peak systolic velocity within the normal physiological range, further confirming the functional feasibility of the graft as a blood conduit.

FIGURE 6.

Evaluation of patency and in vivo tissue regeneration after canine carotid artery transplantation. (a) Ultrasound image one month after canine carotid artery replacement surgery; (b) Ultrasound image of native canine carotid artery; (c) H&E staining, Masson's trichrome staining, and immunofluorescence staining of α‐SMA and CD31 one month after canine carotid artery replacement surgery.

After explantation at one month, H&E staining (Figure 6c) showed that the DBT‐0.2 graft maintained an intact tubular structure with no thrombotic occlusion in the lumen. A large number of cell nuclei were observed infiltrating from the outer to the inner layers of the graft wall, suggesting successful infiltration and migration of host fibroblasts, smooth muscle progenitor cells, and other cell types into the three‐dimensional network of the PET/PGA fiber scaffold. This host cell invasion follows the spatial pathways originally created by the degrading PGA sheath, consistent with the muscle‐mimetic design where the “sheath” degrades to allow tissue ingrowth while the “core” remains as a guiding track.

Masson's trichrome staining (Figure 6c) revealed abundant blue‐stained collagen fibers deposited not only in the outer layer but also deep within the medial portion of the graft wall. This newly formed collagen matrix is critical evidence of tissue regeneration and mechanical strength restoration, suggesting that the recruited host cells are highly active and are secreting extracellular matrix (ECM). This robust host repair response is likely triggered and facilitated by the degradation products of PGA and the residual ECM components. Although not yet continuous, positive CD31 signals (red) were clearly visible along the luminal surface and within the graft wall (Figure 6c), demonstrating the onset of endothelial cell adhesion, proliferation, and migration.

This process is essential for the long‐term patency of small‐diameter vascular grafts, as a functional endothelial layer serves as a natural barrier against coagulation and hyperplasia. Meanwhile, widespread α‐SMA (green) expression was observed throughout the graft wall (Figure 6c), indicating that a significant proportion of the recruited cells had differentiated into contractile smooth muscle cells (SMCs). Such guided differentiation recapitulates the developmental cues present in hierarchical muscle tissue, where the topological and mechanical microenvironment directs progenitor cell fate. These cells, which are major components of the vascular medial layer and are responsible for vasoconstriction and mechanical support, signify progression toward more mature vascular tissue regeneration.

The long‐term presence of the PET framework warrants discussion. Although PET is a non‐degradable polymer, the use of fine multifilaments as a reinforcing core offers a unique advantage in this design. The PET fibers are completely encapsulated by the PGA sheath, and as the PGA degrades, they become embedded in the newly regenerated host tissue, forming a reinforced composite vascular wall. This encapsulation minimizes the risk of foreign body reactions and compliance mismatch, as the thin, flexible PET filaments provide mechanical strength without compromising the graft's compliance. Future long‐term animal studies will be conducted to further evaluate the long‐term safety and remodeling process of the PET‐reinforced graft.

Collectively, these results demonstrate that the muscle‐mimetic core‐sheath strategy enables staged vascular regeneration: the PGA sheath degrades to create space for host cell infiltration, while the PET core provides long‐term mechanical support, and the ECM deposited during subcutaneous implantation promotes rapid endothelialization and smooth muscle differentiation after grafting.

4. Conclusions

Inspired by the hierarchical architecture of skeletal muscle, we developed a novel small‐diameter TEVG based on a muscle‐mimetic core‐sheath composite yarn. The PET/PGA composite yarn, produced via friction spinning technology, successfully integrated the long‐term mechanical strength of PET (as an “artificial myofibril” core) with the bioactivity and degradability of PGA (as a muscle fiber‐like sheath), recapitulating the “basic unit to functional tissue” assembly paradigm of natural muscle. This core–sheath architecture enabled staged and coordinated vascular regeneration through a “yarn mimicking muscle, fabric transforming into vessel” strategy. Subcutaneous implantation in a rabbit model effectively recruited host cells and stimulated the deposition of native extracellular matrix (ECM) on the synthetic scaffold. Subsequent decellularization yielded a cell‐free, ECM‐rich graft that retained favorable biomechanical properties and exhibited excellent cytocompatibility. In vitro evaluations confirmed outstanding anticoagulation performance, cytocompatibility, and structural durability. In a canine carotid artery replacement model, the graft (DBT‐0.2) demonstrated high patency, physiological blood flow, extensive host cell infiltration, and differentiation. Within one month, endothelialization, smooth muscle regeneration, and collagen deposition were observed, indicating active graft remodeling and integration. This proof‐of‐concept study demonstrates that the muscle‐mimetic core‐sheath strategy achieves promising short‐term patency and host cell infiltration in a canine carotid artery model. However, we acknowledge that one‐month implantation is insufficient to evaluate long‐term graft performance. Future studies with extended implantation periods (e.g., 3–6 months) are necessary to assess long‐term patency, mechanical stability, and remodeling outcomes. Nevertheless, the current results support the feasibility of this bioinspired design for small‐diameter vascular regeneration and warrant further optimization toward potential clinical applications.

Author Contributions

Huawang Zhao: writing – review and editing, writing – original draft, visualization, validation, methodology, investigation, formal analysis, data curation, and conceptualization. Cheng Yu: writing – review and editing, writing – original draft, visualization, validation, methodology, investigation, formal analysis, data curation, and conceptualization. Jincheng Zhang: software, investigation, and data curation. Haofeng Liu: software, investigation, and data curation. Shenao Wang: data curation and supervision. Xianzhen Dong: supervision and project administration. Weilin Xu: supervision. Yan Zhuang: supervision and project administration. Hongjun Yang: writing – review and editing, writing – original draft, supervision, project administration, methodology, and funding acquisition.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adhm71413‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.