Exercise-mimetic electrical stimulation of muscles triggered by sonopiezoelectric therapy promotes functional repair of abdominal wall defects

Abstract

Current clinical approaches for abdominal wall defects (AWD) caused by extensive myofascial tissue loss employ synthetic meshes for fibrotic tissue repair, but fail to restore complete functionality. Under physiological conditions, rhythmic contractions of abdominal wall muscles during respiration generate endogenous bioelectric cues essential for muscle action potentials and cellular behavior. Therefore, we aimed to develop an exercise-mimetic electrical stimulation (ES) triggered by sonopiezoelectric therapy (SPT) to promote functional repair of AWD. To achieve this, a sandwich-like piezoelectric patch (PCP) was fabricated by sandwiching a chitin mat between two polycaprolactone mats. The three-layer structure mimics lateral abdominal muscle microarchitecture and provides topographical cues for myoblast alignment and fusion. Under respiratory-induced mechanical deformation, the PCP patches generate ES, helping restore local electric field homeostasis. Additionally, low-intensity pulsed ultrasound (LIPUS) can be applied to deliver more precise and intensified stimulation, further promoting myogenic differentiation and myoblast fusion in both in vitro and in vivo studies. It was found that ES could activate the Ca²⁺/CaN/NFAT2 signaling pathway under dual mechanical and LIPUS stimulation, enhancing the functional repair of AWD. This exercise-mimetic ES triggered by SPT establishes an innovative therapeutic strategy for abdominal wall repair, demonstrating broad potential in regenerative medicine.

Graphical abstract

The exercise-mimetic electrical stimulation system comprises a sandwich-like piezoelectric patch constructed from aligned PCL-chitin-PCL nanofiber mats. It converts both endogenous and exogenous mechanical stimuli into electrical signals, thereby restoring local electric field homeostasis and promoting muscle differentiation and regeneration. The resulting bioelectrical cues act synergistically with the anisotropic topography to guide myofiber alignment and enhance myogenic differentiation via the Ca²⁺/CaN/NFAT2 pathway, ultimately facilitating functional muscle regeneration and reducing fibrosis in abdominal wall defect repair.

1. Introduction

The abdominal wall serves as the primary outer structure of the human abdomen, playing a crucial role in protecting the abdominal organs and maintaining vital functions. Abdominal wall defects (AWD) are common and serious surgical condition, typically resulting from factors such as traumas, soft tissue infections and tumor excision [1]. Surgical intervention utilizing synthetic or biological materials is currently the primary strategy for addressing AWD, aiming to protect abdominal organs while promoting tissue regeneration and functional repair through the deployment of implantable materials [2]. However, implanted materials typically act as foreign bodies, eliciting non-specific inflammatory responses that may result in complications such as infections, enteric fistulas, and intestinal obstructions [3]. Annually over 20 million abdominal wall repair surgeries are performed worldwide using hundreds of implant materials [4]. However, none of these materials has achieved optimal repair of AWD (see Scheme 1).

Scheme 1.

Schematic illustration of the exercise-mimetic electrical stimulation system based on a sandwich-like piezoelectric patch (PCP) and triggered by sonopiezoelectric therapy, highlighting its roles and underlying mechanisms in the functional repair of abdominal wall defects.

AWD are often characterized by substantial loss of myofascial tissue, with muscle regeneration and fibrosis exhibiting a dynamic antagonistic interaction during the repair process [5]. Severe injury or suboptimal microenvironments impair muscle satellite cell (MuSCs)-mediated regeneration, triggering aberrant fibroblast proliferation and excessive collagen deposition that form rigid scars [5]. Current clinical meshes promote fibrotic encapsulation for rapid closure but fail to restore muscle architecture or biomechanical integrity [6], resulting in high recurrence rates of 10%–15% [7]. To overcome the above limitations, emerging tissue engineering strategies have introduced drug delivery systems, and 3D printing technologies to modulate immune responses and enhance mechanical compatibility [8,9]. However, these approaches largely focus on structural restoration and early-stage healing, while overlooking the critical need for functional regeneration. Therefore, the development of implant materials capable of inducing myogenic differentiation is pivotal for achieving functional repair of AWD [10].

It is worth noting that electrical stimulation (ES) play an important role in muscle function maintenance. Skeletal muscle cells generate intrinsic electric fields through membrane potentials and are highly sensitive to ES [11]. The application of external currents with specific frequencies and pulse parameters can induce action potentials, promote calcium release from the sarcoplasmic reticulum, and trigger myotube contraction, thereby not only activating muscle function but also creating a bioelectrical environment that supports the differentiation and maturation of MuSCs [12]. Under physiological conditions, rhythmic contraction-relaxation cycles of abdominal wall muscles during respiration generate dynamic mechanical stresses and endogenous bioelectric cues, which collectively maintain local electric field homeostasis to support tissue metabolism, self-repair, and functional integrity [13]. However, conventional ES techniques such as direct current or capacitive coupling are limited in clinical application due to their invasive nature and high risk of infection [14]. In contrast, piezoelectric materials can autonomously generate electrical signals in response to mechanical stress, offering a transformative strategy for constructing biomimetic electrical microenvironments [15]. Under the periodic mechanical strain induced by respiratory motion of the abdominal wall, piezoelectric materials can generate localized exercise-mimetic electrical signals that may potentially influence MuSC-mediated regeneration and support functional repair. However, in cases of severe AWD or impaired motor function, spontaneous mechanical deformation may be insufficient to elicit an effective piezoelectric response [16]. To address this limitation, sonopiezoelectric therapy (SPT) can be applied as a noninvasive exogenous stimulus to modulate the electrical output of the material in a controlled manner [17,18], thereby establishing a stable exercise-mimetic electrical environment that may further support MuSC differentiation and maturation, potentially facilitating tissue regeneration.

Despite recent advances in piezoelectric biomaterials for muscle regeneration, conventional options such as lead zirconate titanate, polyvinylidene fluoride, and zinc oxide face significant barriers to clinical translation due to concerns over poor biodegradability and potential cytotoxicity [19,20]. In contrast, naturally derived biocompatible piezoelectric materials—particularly proteins (silk [21]) and polysaccharides (cellulose [22] and chitin [23])—have emerged as promising candidates for implantable scaffolds due to their inherent biodegradability and favorable mechanical properties. Chitin, the second most abundant natural polysaccharide, is mainly derived from the exoskeletons of crustaceans such as shrimp and crabs, with an estimate global annual production exceeding 100 billion tons [24]. Its excellent biocompatibility, inherent antibacterial activity and robust mechanical strength make it widely applicable in biomedical, drug delivery, and food preservation fields [25]. The inherent polarity of functional groups such as amino (-NH₂) and acyl (-COCH₃) moieties within its polymer chains enables piezoelectric behavior under mechanical deformation [26]. Notably, chitin exists in three crystalline forms α, β and γ, among which α-chitin and β-chitin possess intrinsic piezoelectric activity [27]. Studies utilizing aligned chitin nanofibers have reported a piezoelectric output of 0.25 V, underscoring their potential for effective mechanoelectrical transduction [26]. Given the anisotropic nature of abdominal wall muscles, aligned piezoelectric nanofiber mats fabricated by electrospinning can not only guide MuSCs alignment and differentiation through topography-mediated biomechanical cues [28], but also amplify piezoelectric performance through structural optimization [29]. However, the application of chitin-based piezoelectric biomaterials for the functional repair of AWD remains largely unexplored and represents an emerging frontier.

Herein, an exercise-mimetic ES system triggered by SPT was proposed as a therapeutic strategy to promote the functional repair of AWD. To achieve this, a sandwich-like piezoelectric patch (PCP) composed of chitin and PCL was fabricated via electrospinning, with PCL included to provide mechanical compliance, enhance fatigue resistance, and maintain the structural integrity of the patch. Under respiratory-induced rhythmic mechanical deformation, the system is capable of generating localized exercise-mimetic ES to help restore local electric field homeostasis. Furthermore, exogenous low-intensity pulsed ultrasound (LIPUS) could be applied to deliver more precise, controllable, and intensified stimulation, thereby further enhancing muscle differentiation and regeneration. Moreover, the aligned fibrous architecture provides topographical cues that guide muscle fiber alignment and mimic the native muscle structure. The mechanical properties, biocompatibility and piezoelectric properties of PCP will be detected. The regulatory effects of this exercise-mimetic ES system on in vitro myogenic differentiation of MuSCs will be evaluated, and the underlying molecular mechanism will be explored. Furthermore, a full-thickness AWD model in rats was constructed to evaluate its in vivo reparative efficacy.

2. Results and discussion

2.1. Characterizations of nanofiber mats

Scanning electron microscope (SEM) was employed to characterize the size and morphology of the nanofiber mats fabricated with varying concentrations of chitin and polycaprolactone (PCL) (Fig. 1A). The nanofiber diameter increased with chitin concentrations, with diameters of approximately 1.5 ± 0.3 μm, 2.1 ± 0.3 μm, and 3.0 ± 0.2 μm for mats containing 0.5 %, 1 %, and 1.5 % chitin, respectively (Fig. 1B). In contrast, the 15 % PCL nanofiber mats exhibited a smaller average diameter of about 0.23 ± 0.04 μm. Additionally, the SEM images revealed an aligned arrangement of all the nanofibers.

Fig. 1.

(A) SEM image (scale bar = 20 μm) and (B) Diameter distribution of the nanofibers. (C) FTIR spectra. (D) XRD spectra. (E) Water contact angle analysis. ∗, p < 0.05.

Chitin naturally exists in three crystalline forms: α, β, and γ, with β-chitin known for its consistent molecular chain arrangement and enhanced piezoelectric properties [27]. Given the influence of acidic solvents, high-voltage conditions during electrospinning, post-spinning heat treatment, and the highly oriented fiber structure induced by alignment-oriented spinning, crystalline transformations of chitin are expected [30]. To investigate these potential changes in the crystalline structure, Fourier transform infrared spectroscopy (FTIR) and X-ray diffraction (XRD) analyses were performed. α-chitin is characterized by a double peak in the amide I region (1625 and 1660 cm⁻¹), whereas β-chitin exhibits a single peak around 1625 cm⁻¹ [31]. In the FTIR spectrum (Fig. 1C) of the chitin powder, the amide I double peaks at 1625 cm⁻¹ and 1660 cm⁻¹ were clearly observed, confirming the presence of α-chitin. However, in the chitin nanofiber mats, the 1660 cm⁻¹ amide I peak was less pronounced, suggesting a possible shift toward the β-crystalline form. XRD analysis (Fig. 1D) provided further evidence of this transformation, as indicated by a rightward shift in the (020) diffraction peak and the disappearance of the 23.4° reflection, both confirming the β-transformation of the chitin crystalline structure [32]. The presence of abundant amino (-NH₂) and hydroxyl (-OH) functional groups in chitin imparts hydrophilic properties [33]. Contact angle analysis (Fig. 1E) revealed that both chitin and PCL mats exhibited moderate hydrophilicity (<90°), with chitin mats (∼54°) demonstrating stronger hydrophilic behavior than PCL mats (∼87°).

2.2. Mechanical and piezoelectric properties of the nanofibrous mats

The mechanical properties of the nanofibrous mats were showed in Fig. 2A. The abdominal wall muscles experiences approximately 3–4% bending deformation during respiration [34], and an ideal repair material should withstand the maximum abdominal pressure (∼0.02 MPa) [35]. The maximum tensile strength of the commonly used abdominal wall repair materials is approximately 20 N/cm [36], corresponding to 2 MPa for a 1 mm-thick patch. The nanofiber mats with chitin concentrations of 0.5 %, 1 %, and 1.5 % exhibited tensile stress of 6.18 ± 0.14 MPa, 9.45 ± 0.13 MPa, and 13.08 ± 0.20 MPa, respectively (Fig. 2B), meeting the mechanical strength requirements for abdominal wall repair materials. The 1 % chitin nanofiber mats, exhibiting the highest elongation at break, were selected for further experiments. The tensile-recovery curves shown in Fig. 2C illustrated that the elongation is fully reversible after 50 repeated loading-unloading cycles. Consequently, chitin nanofiber mats can effectively accommodate the synchronized tensile and recovery deformations induced by abdominal wall respiration during regenerative repair.

Fig. 2.

Mechanical and piezoelectric properties of the nanofibrous mats. (A) Typical stress-strain curves. (B) Tensile stress. (C) Tensile-recovery curves of chitin nanofibrous mats. (D) Piezoelectric output of chitin nanofibrous mats with different chitin concentrations. (E) Piezoelectric output of chitin nanofibrousmats under LIPUS stimulation at different intensities. (F) Schematic representation of the construction of the sandwich-like patch. ∗, p < 0.05.

Bending mechanical stimulation was applied in piezoelectric measurements to simulate respiratory strain of the abdominal wall. The results of Fig. 2D demonstrated that higher chitin concentrations enhanced piezoelectric performance, with the 1 % chitin nanofiber mats producing an output voltage of up to 0.09 V, and the 1.5 % mats generating up to 0.3 V. To evaluate the ultrasonic response, the nanofiber mats were subjected to LIPUS stimulation at varying intensities (1 MHz), with the results indicating an increase in output voltage as the LIPUS intensity was elevated (Fig. 2E). To better simulate the physiological environment of abdominal wall defect repair, a piezoelectric platform was constructed using the PCP patch, aluminum foil, copper wires, and polydimethylsiloxane (PDMS) encapsulation layers. Immersed in PBS 2 cm from LIPUS (1 W/cm²), the patch showed stable output at 0.2 V with only mild attenuation (Fig. S1). Given the PCP patch is 1 mm thick, the measured 0.2 V corresponds to a local electric field of ∼0.2 V/mm, falling within the 0.1–0.5 V/mm range [37] which could enhance myotube formation and functional maturation.

To simulate the muscle orientation of the abdominal wall and enhance the functional repair and regeneration, we designed an exercise-mimetic ES system consisting of a sandwich-like structure of PCL-chitin-PCL (PCP), with each layer exhibiting a distinct fiber alignment (Fig. 2F). The orientation of each layer mirrors the native muscle structure of the abdominal wall, corresponding to the arrangement of the transversus abdominis (transverse muscle), internal oblique (lower 45°), and external oblique (upper 45°).

2.3. Cytocompatibility of the exercise-mimetic ES system

Cell viability at 24, 72, and 120 h showed no significant differences among the three groups at any time point (Fig. 3A). The cell survival rate increased over time in all groups. Only a few dead cells were observed, while the number of live cells steadily increased, indicating effective proliferation and spreading on the surface of the PCP mats (Fig. 3B). Additionally, L6 cells exhibited alignment along the orientation of the nanofibers, this observation confirmed that the aligned fiber topography could guide cell growth. Cytoskeletal analysis further corroborated this finding, revealing that the aligned fibers directed cell alignment along the longitudinal axis of the fibers (Fig. 3C). Furthermore, no significant differences in cell morphology were observed on the surfaces of different materials.

Fig. 3.

Cytocompatibility of the exercise-mimetic ES system. (A) CCK8 assay of L6 cells cultured on different materials for 24 h, 72 h and 120 h. (B) Live/dead staining of L6 cells cultured on different materials for 24 h, 72 h and 120 h (scale bar = 200 μm). (C) Cytoskeleton staining by phalloidin (red) of L6 cells cultured on different materials for 72 h (scale bar = 50 μm), nuclei were stained as blue by DAPI. ns, p > 0.05.

2.4. Regulation on myogenic differentiation of L6 cells by the exercise-mimetic ES system

L6 cells can differentiate into multinuclear myotubes in the myogenic medium. To further assess the impact of different materials (PCL and PCP) on in vitro myogenic differentiation of L6 cells, quantitative PCR (qPCR) analysis was performed for key regulatory genes, including PAX7, Myf5, MyoD, and Myo, to evaluate cell activation and progression through early and late stages of differentiation. The results showed no significant differences between PCL and PCP, suggesting that the material composition did not exert a discernible effect on the myogenic differentiation of L6 cells (Fig. S2). Since no significant differences in myogenic differentiation were observed among the materials, we proceeded to investigate the effects of exercise-mimetic ES generated by the PCP patch. qPCR analysis demonstrated that after 3 days of LIPUS stimulation, cells cultured on PCP patches with exercise-mimetic ES (PCP + LIPUS group) showed a significantly elevated expression of PAX7, Myf5, MyoD, and MyoG (Fig. 4A). This increase reflects early activation of L6 cells and the initiation of the myogenic program at the molecular level, preceding the formation of multinucleated myotubes and potentially contributing to subsequent physiologically relevant muscle differentiation. Furthermore, the elevated gene expression compared to the control groups indicates that the enhancement of myogenic differentiation is primarily attributable to the exercise-mimetic ES rather than the ultrasound or material properties.

Fig. 4.

Regulation on myogenic differentiation of L6 cells by the exercise-mimetic ES system. (A) Gene expression detected by qPCR for PAX7, Myf5, MyoD, and MyoG of L6 cells cultured on different materials at 3 days. (B) Western blot analysis for MyoD, MHC and MyoG. (C) Quantitative analysis of Western blot results. (D) Immunofluorescence staining of MyoG and MHC in L6 cells with or without LIPUS stimulation at day 8 (scale bar = 200 μm). ∗, p < 0.05.

To further investigate the impact of exercise-mimetic ES from PCP patches on myotube formation, we additionally measured protein expression of MHC (Myosin Heavy Chain). Consistent with previous qPCR results, protein expression of MyoD, MyoG, and MHC was highest in the PCP + LIPUS group (Fig. 4B), with quantitative analysis revealing statistically significant differences (Fig. 4C). Another important characteristic of myogenic differentiation is cell alignment and cell fusion, which are essential processes associated with myotube formation [38]. As depicted in Fig. 4D, both MyoG and MHC expression levels were significantly upregulated in cells stimulated by the exercise-mimetic ES system compared to the control group. Quantitative analysis further confirmed that this increase was statistically significant (Fig. S3). Therefore, it was demonstrated that LIPUS-triggered exercise-mimetic ES from PCP patch enhanced the expression of genes and proteins involved in myogenic differentiation across both early and late stages.

2.5. Transcriptomic analysis of L6 cells under exercise-mimetic ES

RNA sequencing and transcriptomic analysis were conducted to elucidate the biological effects of the exercise-mimetic ES on L6 cells. The volcano plot (Fig. 5A) revealed significant differences in gene expression between the PCL + LIPUS group and PCP + LIPUS group. As shown in Fig. 5B, the upregulated Differentially Expressed Genes (DEGs) in the PCP + LIPUS group were associated with muscle cell differentiation, response to mechanic stimulus, calcium ion import, angiogenesis, and nerve development. Notably, key upregulated genes, such as CCN2 [39] (Connective Tissue Growth Factor) is essential for extracellular matrix synthesis and remodeling, contributing to muscle tissue recovery after injury by regulating fibrosis and repair processes. MYH4 [40] (Myosin Heavy Chain 4) is associated with the formation of fast-contracting muscle fibers and serves as a key marker of muscle maturation. CACNA1S [41] (Calcium Voltage-Gated Channel Subunit Alpha1 S) plays a critical role in regulating calcium channels in muscle cells, which are vital for muscle contraction and mechanical activity. Additionally, ITPRIP [42] (Inositol 1,4,5-Trisphosphate Receptor Interacting Protein) plays a role in regulating calcium signaling, while P2RX6 [43] (Purinergic Receptor P2X6) influences ion channel function, both contributing to the modulation of muscle cell activity and differentiation. In contrast, ID2 [44] (Inhibitor of DNA Binding 2) and ID4 [44] (Inhibitor of DNA Binding 4) act to inhibit muscle differentiation and preserving their proliferative capacity. Their downregulation promotes muscle differentiation by relieving this inhibition.

Fig. 5.

Transcriptomic analysis of L6 cells under exercise-mimetic ES. (A) Volcano plot of differential expression genes between PCL + LIPUS group and PCP + LIPUS group. (p < 0.05 and fold change >2). (B) Heatmap of genes related to muscle cell differentiation, response to mechanic stimulus, calcium ion import, angiogenesis, and nerve development between PCL + LIPUS group and PCP + LIPUS group. (p < 0.05 and Fold change >2). (C) GO analysis of genes upregulated in PCP + LIPUS group than PCL + LIPUS group. (p. adj <0.05). (D) Scheme diagram of the possible mechanism of exercise-mimetic ES in regulating L6 cell function.

Furthermore, Gene Ontology (GO) analysis revealed that the PCP + LIPUS group exhibited more upregulated genes associated with processes such as positive regulation of release of sequestered calcium ion into cytosol, cell adhesion, positive regulation of cell population proliferation, positive regulation of angiogenesis, extracellular matrix organization, positive regulation of vascular endothelial growth factor production, response to mechanical stimulus, angiogenesis, muscle cell differentiation, myoblast differentiation, regulation of calcium ion transmembrane transport, establishment or maintenance of transmembrane electrochemical gradient, positive regulation of actin filament bundle assembly, regulation of calcium-mediated signaling, muscle myosin complex, nerve development, intracellular calcium ion homeostasis, positive regulation of calcium ion import, skeletal system development, and myosin filament assembly, compared to the PCL + LIPUS group (Fig. 5C).

Interestingly, our analysis revealed that the enriched processes are primarily associated with calcium ion signaling. Previous studies have corroborated similar findings [45]. Exercise-mimetic ES induced by the piezoelectric effect alters the membrane potential (V_mem) of L6 myoblasts [46], leading to cellular depolarization and activation of voltage-gated calcium channels. This causes a transient increase in intracellular calcium ion concentration, which subsequently activates several downstream signaling pathways [47], including the Ca²⁺/CaM (Calmodulin)-dependent phosphatase CaN (Calcineurin)/NFAT2 (Nuclear Factor of Activated T cells) pathway, the Ca²⁺/CaMK (Calmodulin-dependent Kinase)/MEF2 (Myocyte Enhancer Factor 2) pathway, and the Ca²⁺/MAPK (Mitogen-Activated Protein Kinase)/cyclin D1 pathway. Collectively, these signaling pathways regulate the expression of differentiation-related genes, thereby promoting myogenic differentiation and myoblast fusion [48].

Consequently, we investigated the Ca²⁺ signaling pathway. Western blot validation was performed, confirming the upregulation of NFAT2 (Fig. 6G). Thus, a postulation was made that LIPUS-induced exercise-mimetic ES from the piezoelectric PCP patches modulates myoblast V_mem, thereby initiating activation of voltage-gated calcium channels (CACNA1S). This cascade stimulates the Ca²⁺/CaN/NFAT2 signaling axis, facilitating NFAT2 dephosphorylation and subsequent nuclear translocation. These signaling mechanisms ultimately orchestrate transcriptional regulation of downstream targets (MyoD, CCN2), effectively coordinating myogenic differentiation and ECM remodeling processes (Fig. 5D).

Fig. 6.

Role of the Ca²⁺/CaN/NFAT2 signaling pathway in myogenic differentiation induced by exercise-mimetic ES. (A) Detection of intracellular Ca²⁺ concentration of L6 cells. (B) Detection of intracellular Ca²⁺ concentration of L6 cells by Rhod-2/AM fluorescence staining (scale bar = 100 μm). (C) Quantitative analysis of Rhod-2/AM fluorescence staining. (D) Immunofluorescence staining of the CACNA1S channel in L6 cells (scale bar = 100 μm). (E) Quantitative analysis of immunofluorescence staining of the CACNA1S channel. (F) Immunofluorescence staining of the intranuclear translocation of NFAT2 in L6 cells (scale bar = 50 μm). (G) Western blot analysis for CaN, NFAT2, p-NFAT2, MyoD, and CCN2 in L6 cells with or without verepamil cultured on different samples at day 4. (H) Quantitative analysis of Western blot analysis. Nuclei were stained as blue by DAPI. ∗, p < 0.05.

2.6. Role of the Ca²⁺/CaN/NFAT2 signaling pathway in myogenic differentiation induced by exercise-mimetic ES

To investigate the role of Ca²⁺ influx in myogenic differentiation induced by exercise-mimetic ES, verapamil (100 μM) was added to the medium as a blocker of voltage-gated calcium channels. All samples, with or without verapamil treatment, were collected on day 4. Intracellular Ca²⁺ concentration was then assessed. As depicted in Fig. 6A, the intracellular Ca²⁺ concentration was significantly elevated in the PCP + LIPUS group compared to the PCL + LIPUS group, in the absence of verapamil, whereas no significant difference was observed between the two groups after verapamil treatment. This suggests that exercise-mimetic ES facilitated Ca²⁺ influx in L6 cells and increased intracellular Ca²⁺ concentration which was not triggered by LIPUS stimulation. To further visualize the Ca²⁺ influx induced by exercise-mimetic ES through piezoelectric PCP patches, Rhod-2/AM as a Ca²⁺ probe was utilized for fluorescence staining. The results (Fig. 6B) further corroborated that the intracellular Ca²⁺ fluorescence intensity in the PCP + LIPUS group was markedly higher compared to the PCL + LIPUS group without verapamil. Quantitative analysis revealed that this difference was statistically significant (Fig. 6C). Meanwhile, a downward trend in Ca²⁺ fluorescence density was observed in the PCL + LIPUS group following verapamil treatment, indicating potential Ca²⁺ influx induced by LIPUS stimulation. However, the Ca²⁺ influx was still primarily driven by the exercise-mimetic ES generated by the piezoelectric system.

Further investigation was conducted to identify the specific voltage-gated calcium channels involved in mediating Ca²⁺ influx induced by exercise-mimetic ES. Although RNA sequencing identified several differentially expressed calcium channels, CACNA1S, a voltage-gated L-type calcium channel predominantly expressed in skeletal muscle cells, plays a critical role in modulating cellular electrical activity and excitability in response to voltage fluctuations, thereby serving as the primary subject of our investigation. To assess its expression, immunofluorescence staining was then performed. As shown in Fig. 6D, CACNA1S expression was significantly upregulated in cells from the PCP + LIPUS group compared to those from the PCL + LIPUS group, an effect that was subsequently attenuated by verapamil. Quantitative analysis further confirmed the statistical significance of the observed results (Fig. 6E). These findings suggest that exercise-mimetic ES promotes Ca²⁺ influx not only by activating voltage-gated calcium channels but also through the upregulation of voltage-gated calcium channel protein expression.

After confirming that exercise-mimetic ES facilitates intracellular Ca²⁺ influx through CACNA1S and initiates downstream signaling cascades, we further investigated the specific signaling pathways involved in piezoelectric stimulation. Among the calcium channel-related signaling pathways, the NFAT47] family plays a pivotal role in skeletal muscle development and regeneration. As intracellular Ca²⁺ increases, binds to CaM, activates phosphatase CaN, dephosphorylates NFAT, and induces its translocation into the nucleus, where it subsequently regulates the transcription of target genes, ultimately modulating muscle cell differentiation and regeneration [49]. We first assessed the expression of related proteins involved in this signaling pathway through Western blot analysis. The results (Fig. 6G) revealed that in the absence of verapamil, the PCP + LIPUS group exhibited significantly elevated expression levels of CaN, NFAT2, MyoD, and CCN2 compared to the PCL + LIPUS group, while p-NFAT2 expression was notably reduced. Upon verapamil treatment, the expression of these proteins was restored, and the quantitative analysis demonstrated statistical significance (Fig. 6H). To further visualize the intranuclear translocation of NFAT2, immunofluorescence staining was performed. The results (Fig. 6F) corroborated the Western blot findings and showed that the exercise-mimetic ES significantly facilitated NFAT2 nuclear translocation while verapamil effectively inhibited this process.

Collectively, these results indicate that piezoelectric PCP patches under LIPUS stimulation generate localized electrical cues that likely activate CACNA1S channels and elevation of intracellular Ca²⁺ levels. This Ca²⁺ increase is accompanied by activation of the Ca²⁺/CaM-dependent CaN pathway and enhancement NFAT2 nuclear translocation, together with upregulation expression of myogenic regulators such as MyoD and CCN2. Taken together, these findings support a calcium-dependent mechanistic framework through which piezoelectric stimulation may promote myogenic signaling.

2.7. Effects of the exercise-mimetic ES system on abdominal wall defect repair in rats

A full-thickness abdominal wall defect model (2 × 1 cm²) was established on the lateral abdominal wall of rats, with PCL and PCP patches sutured to the defect site using the inlay surgical technique. Fig. 7A illustrates the process of establishing the abdominal wall defect model. The experiment was divided into four groups: the normal group (uninjured abdominal wall), the PCL group, the PCP group, and the PCP + LIPUS group. The repair and regeneration processes of abdominal wall defects were observed, sampled, and evaluated at 4 and 8 weeks post-operation. Throughout the experiment, no adverse events including bleeding, infection, tissue overheating, tissue damage, or mortality were observed.

Fig. 7.

In vivo abdominal wall defect repair. (A) Schematic diagram of the abdominal wall defect model and surgical procedure. (B) Morphology of the abdominal wall defects muscle at week 4 and 8 postoperative. (C) Representative images of H&E staining and Masson staining of abdominal wall defects muscle treated in different groups at week 4 postoperative (scale bar = 200 μm, 50 μm). (D) Representative images of H&E staining and Masson staining of abdominal wall defects muscle treated in different groups at week 8 postoperative (scale bar = 200 μm, 50 μm). (E) Quantitative analysis of collagen volume fraction. (F) Quantitative analysis of myofiber diameters. ∗, p < 0.05.

The appearance of the repaired abdominal wall muscles is shown in Fig. 7B. By the fourth week, the original defects in the PCL, PCP, and PCP + LIPUS groups were filled with regenerated tissue. However, the remnants of the patches could still be viewed in all groups, with the PCL group showing the most prominent presence, indicating that the regenerated tissue was relatively weak. By the eighth week, in the PCL group, due to the insufficient mechanical support, the regenerated abdominal wall tissue was thinner and exhibited slight bulging under abdominal pressure. Substantial degradation of the patches was observed in both the PCP and PCP + LIPUS groups, with the regenerated muscle at the degraded sites exhibiting a satisfying appearance. These findings suggest that the PCP patches, by harnessing the piezoelectric effect induced by abdominal wall respiratory strain and sonopiezoelectric therapy, further enhanced muscle tissue regeneration.

The morphology of regenerated muscle was further evaluated in the abdominal wall defect model using H & E and Masson's trichrome staining. At 4 weeks postoperatively (Fig. 7C), significant local inflammation and muscle loss were observed at the defect sites in the PCL, PCP, and PCP + LIPUS groups, with noticeable patch remnants (Fig. S4). Notably, collagen accumulation (stained as blue) was less pronounced in the PCP and PCP + LIPUS groups, while muscle regeneration (stained as red) was more prominent compared to the PCL group. Quantitative analysis of transversus abdominis myofiber orientation revealed that regenerated muscle fibers in the PCL, PCP, and PCP + LIPUS groups all exhibited a certain degree of directional alignment. This is consistent with the guidance effect provided by the aligned nanofiber architecture of the scaffolds (Fig. S5). The collagen volume fraction (CVF) was quantitatively measured to assess the extent of fibrosis within the regenerated tissue (Fig. 7E). The results revealed that the CVF in the PCL group was significantly higher than in the PCP and PCP + LIPUS groups, suggesting that the PCP patches facilitated muscle regeneration while reducing collagen deposition and fibrosis more effectively. However, no significant difference in CVF was observed between the PCP and PCP + LIPUS groups at this stage. This lack of difference may be attributed to the short duration of the repair period, indicating that the additional exercise-mimetic ES had not yet exerted a notable impact on muscle regeneration. A quantitative analysis of myofiber diameters was then performed (Fig. 7F). Although showing lower myofiber diameter than normal group, better organization and higher diameter of myofibers could be viewed in the PCP and PCP + LIPUS groups, and statistical analysis showed these groups had significantly higher myofiber diameters than the PCL group.

At 8 weeks postoperatively (Fig. 7D), the degradation of PCL patches led to a loss of mechanical integrity, resulting in a sustained fibrotic process within the defect area that suppressed muscle regeneration and formed extensive fibrotic scar tissue. In contrast, both the PCP and PCP + LIPUS groups exhibited pronounced regenerative efficacy, characterized by larger myofiber diameters and lower CVF compared to the 4-week time point. Consistently, quantitative analysis revealed a decline in myofiber orientation coherence in the PCL group relative to week 4, whereas the PCP and PCP + LIPUS groups maintained robust and coherent alignment within the transversus abdominis layer, indicating sustained anisotropic muscle regeneration (Fig. S5). Notably, the PCP + LIPUS group demonstrated a significantly higher quantity of regenerated muscle fibers and greater fiber diameters than the PCP group at 8 weeks. This result indicates that the additional exercise-mimetic ES provided by LIPUS exerted a synergistic effect in promoting long-term muscle regeneration. Furthermore, the diminishing disparity between regenerated muscle tissue in the PCP + LIPUS group and native abdominal wall musculature suggests that the synergistic action of respiratory strain-induced piezoelectric stimulation and sonopiezoelectric therapy-enhanced electrical cues augmented the regenerative potential of the exercise-mimetic ES system for abdominal wall defect repair.

To clarify whether LIPUS exerts effects independent of the piezoelectric patch, an additional LIPUS group was included at the 2-week time point. Histological analysis (Fig. S6) showed that LIPUS alone induced minimal myogenic tissue compared with the PCP + LIPUS group, indicating that LIPUS alone did not induce appreciable muscle regeneration. This observation is consistent with previous studies reporting the limited regenerative efficacy of LIPUS in the absence of piezoelectric activation or additional electromechanical cues [50,51].

The maturity and contractile functionality of regenerated muscle was further evaluated using MHC immunofluorescence staining (Fig. 8A and B), accompanied with subsequent quantitative analyses. Results demonstrated that both the PCP and PCP + LIPUS groups exhibited superior myofiber maturation and contractile properties compared to the PCL group at both time points. Additionally, the PCP + LIPUS group displayed progressively enhanced myofiber maturation and contractile capacity over time, with improvements being more pronounced. Quantitative analyses confirmed significant differences among these groups. Biomechanical test was conducted on the regenerated muscle to evaluate functional recovery. Quantitative analysis indicated that all groups exhibited reduced tensile strength at week 4, with the PCL group showing the highest Young's modulus. By 8 weeks, tensile strength showed marked improvement across all groups, with the PCP + LIPUS group approaching normal levels. Nevertheless, the PCL group maintained the highest Young's modulus, likely due to predominant rigid scar tissue formation (Fig. S7).

Fig. 8.

Representative images of immunofluorescence staining and statistical analysis of (A, B) MHC, (C, D) CACNA1S, and (E, F) CCN2 in different groups at 4 and 8 weeks. MHC, CACNA1S and CCN2 were stained as green, and nuclei were stained as blue by DAPI. ∗, p < 0.05.

Concurrently, to elucidate the role of the Ca²⁺/CaN/NFAT2 pathway identified in abdominal wall muscle regeneration, immunofluorescence staining of CACNA1S, NFAT2, and CCN2 was performed. Results revealed that the expression of CACNA1S, a voltage-gated calcium channel, was elevated in the PCP + LIPUS group compared to other groups, with quantitative analyses confirming statistical significance (Fig. 8C and D). Notably, CACNA1S expression was exclusively localized to skeletal muscle. While NFAT2 expression showed no marked intergroup differences, this observation aligns with its functional reliance on dephosphorylation-mediated nuclear translocation, a process that cannot be accurately detectable by conventional tissue immunofluorescence staining (Fig. S8). CCN2 expression (Fig. 8E and F) was predominantly observed in the extracellular matrix, with the PCP + LIPUS group exhibiting the highest levels at 4 weeks, suggesting its early-stage regulatory role in modulating extracellular matrix synthesis and degradation, as well as facilitating muscle cell proliferation and differentiation to promote muscle regeneration. However, no significant differences in CCN2 expression were detected among the PCL, PCP, and PCP + LIPUS groups at 8 weeks. Literature review indicates that CCN2 potentiates myogenic differentiation during early phases but may exert inhibitory effects or promote skeletal muscle fibrosis in late stages through overexpression [52]. This biphasic behavior parallels our findings, implying that exercise-mimetic ES regulates abdominal wall repair via the Ca²⁺/CaN/NFAT2 pathway across distinct regenerative phases. In conclusion, these results suggest that the exercise-mimetic ES system can effectively promote functional repair of abdominal wall defects, closely approaching the level observed in the normal group.

3. Conclusion

Abdominal wall defects, particularly those involving extensive myotendinous tissue loss, pose significant clinical challenges due to the lack of ideal repair materials for functional repair and regeneration. In this study, we developed an exercise-mimetic ES system triggered by SPT to enhance myogenic differentiation. The system is based on a sandwich-like piezoelectric patch, which is a three-layer, anisotropically aligned linear piezoelectric nanofiber structure and fabricated from chitin and PCL via electrospinning. It is capable of convert both endogenous and exogenous mechanical stimuli into electrical signals, restoring local electric field homeostasis and further enhancing muscle differentiation and regeneration. Structurally, the oriented nanofibrous patches deliver contact guidance through their anisotropic topography, directing parallel alignment of neo-myofibers to recapitulate native muscle structure. This synergy between the bioelectric and topography cues promotes the functional regenerative repair of abdominal wall defects. In vitro studies demonstrated that the PCP patches significantly enhance myogenic differentiation and myoblast fusion via the Ca²⁺/CaN/NFAT2 pathway upon LIPUS activation. In vivo experiments further validated that the exercise-mimetic ES system effectively promotes abdominal wall muscle tissue regeneration and mitigates fibrosis. The integration of LIPUS further amplified long-term regenerative outcomes by dynamically modulating electrical cues, highlighting the system's adaptability throughout the repair process. In conclusion, the exercise-mimetic ES system exhibits good potential for functional repair of abdominal wall defects, demonstrating broad potential in regenerative medicine.

4. Materials and methods

4.1. Electrospinning of nanofibrous mats

Practical-grade chitin (Perfemiker, China) solutions with concentrations of 0.5 %, 1 %, and 1.5 % (wt/v) were prepared in 1,1,1,3,3,3-hexafluoro-2-propanol (99.5 %, HFIP, Yuanye, China) and continuously agitated at room temperature for 3 days. PCL (Mw = 80,000, Wanvi, China) was dissolved in HFIP at 15 % (wt/v) and stirred overnight at room temperature. All solutions were prepared with materials as received and mixed until fully homogeneous. Nanofibrous mats were fabricated using an electrospinning apparatus (SS-X3, Yongkang Leye, China). The chitin solution was electrospun under conditions of 15 kV, with a distance of 12 cm between the needle and the collector, and a flow rate of 1.0 mL/h. The PCL solution was electrospun at 12 kV, 10 cm, and a flow rate of 0.1 mL/h. Both solutions were processed at 23 °C and 21 % relative humidity, with the collector speed set to 2000 rpm. After electrospinning, the mats were removed and placed in a desiccator overnight for drying.

4.2. Characterizations of the nanofiber mats

We employed field emission scanning electron microscope (SEM) (FEI, TF20, Sigma, Germany) to examine the chitin and PCL nanofibers' morphology. Ultima IV X-ray diffractometer (Rigaku, Japan) was used to analyze both chitin powder and nanofibers' X-ray diffracyion (XRD). Nicolet iS10 spectrometer (Thermo Fisher, USA) was employed to test samples’ Fourier transform infrared spectroscopy (FTIR) spectra. Water contact angle analysis of the chitin and PCL nanofiber mats was conducted using an OCA20 goniometer (Dataphysics, Germany).

4.3. Mechanical test

Chitin and PCL nanofiber mats, measuring 30 × 10 mm², were subjected to tensile testing using a universal testing machine (5982, Instron, USA). Additionally, cyclic tensile tests with a strain amplitude of 5 % were performed on the chitin nanofiber mats to evaluate their fatigue resistance under repeated loading conditions.

4.4. Piezoelectric performance test

Aluminum foil was used to cover the top and bottom surfaces of the nanofiber mats as electrodes. Copper wires were then attached to the aluminum foils to function as electrode leads. The nanofiber mats were subjected to a 10 % bending strain at 1 Hz to simulate the respiratory strain of the abdominal wall. low-intensity pulsed ultrasound (LIPUS) stimulation at 1 MHz with intensities of 0.1, 0.5, and 1 W/cm² was applied to investigate the ultrasonic response of the mats. The sensors at the top and bottom of the sample had a fixed area of 100 mm² (diameter, 10 mm). The leads were connected to an oscilloscope and a digital multimeter (6514, Keithley, USA) for detection and recording of the material's piezoelectric performance.

4.5. Construction of the sandwich-like PCP patch

Each layer of the patch consisted of a 1 × 1 cm² film and 300 μL of rat tail collagen. The patch was assembled in a PCL-chitin-PCL (PCP) sequence, with each film layer corresponding to the orientation of the abdominal wall muscles. Specifically, the layers followed the arrangement of the abdominal muscles: transversus abdominis (transverse), internal obliquus abdominis (internal, lower 45^o), and external obliquus abdominis (external, upper 45^o).

4.6. Cell culture and differentiation of L6 cells

The L6 rat myoblast cell line was obtained from Cell Bank of Chinese Academy of Sciences (China), and the complete medium consisted of Dulbecco's Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F12; BasalMedia, China), 10 % fetal bovine serum (FBS; Gibco, USA), and 1 % penicillin/streptomycin (P/S; BasalMedia). Cultivation was performed at 37 °C under a humidified atmosphere containing 5 % CO₂. Upon reaching approximately 80 % confluence, L6 cells were dissociated using 0.25 % trypsin-EDTA (BasalMedia) and subsequently passaged at a 1:3 ratio.

To induce myogenic differentiation, L6 cells were cultured in differentiation medium—DMEM (BasalMedia) supplemented with 2 % horse serum (Biosharp, China) and 1 % P/S (BasalMedia)—replacing the standard complete medium. The differentiation medium was renewed each day.

4.7. In vitro cytocompatibility

To evaluate the biocompatibility of the exercise-mimetic ES system, L6 myoblasts were seeded onto the surfaces of PCL and PCP nanofiber mats, with cells cultured on Petri dishes as the Control group. Cell Counting Kit-8 (CCK-8; Beyotime, China) was used to evaluate cell viability at 24, 72, and 120 h. Using a microplate reader, absorbance at 450 nm was measured. Cell viability was calculated with the following formula: Cell viability = [OD (sample)-OD (blank)]/[OD (control) − OD (blank)] × 100 %. Additionally, a LIVE/DEAD assay (R37601, Thermo Fisher) was performed at the same points, and the cells were imaged using fluorescence microscopy. The cytoskeleton was stained with rhodamine-labeled phalloidin on day 3. Fluorescence microscope (Zeiss, Germany) was used to inspect stained samples.

4.8. In vitro LIPUS stimulation

Cells were subjected to LIPUS stimulation using an LIPUS physiotherapy device (Dongdixin, China). At a density of 5 × 10⁴ cells/cm², the cells were seeded on PCL and PCP nanofiber mats. LIPUS stimulation was applied after 24 h of incubation with the following parameters: frequency of 1 MHz, power density of 1 W/cm², duty cycle of 20 %, and 5-min exposures every 12 h [53].

4.9. Quantitative PCR analysis

Following equivalent treatment durations across experimental cohorts, total RNA was harvested from L6 myoblasts via TRIzol-based extraction (Invitrogen, USA). Purified RNA underwent reverse transcription into cDNA templates using the iScript system (Bio-Rad, USA). For quantitative PCR (qPCR), cDNA aliquots were combined with sequence-specific oligonucleotides and SYBR Green-based amplification reagents (Bio-Rad), and analyzed using a PCR platform (Applied Biosystems). Transcript abundance for myogenic regulators (PAX7, Myf5, Myog, Myod) was computationally derived, with GAPDH as the endogenous normalization standard. Primer sequences are detailed in Supplementary Primer Repository (Table S1).

4.10. Immunofluorescence staining of L6 cells related to myogenic differentiation

L6 cells were cultured in different groups with myogenic medium for 4 and 8 days. Post-culture specimens underwent sequential pretreatment: PBS rinses, stabilization in 4 % paraformaldehyde (PFA), and membrane solubilization via 0.5 % Triton X-100. Nonspecific epitope saturation was achieved through 5 % bovine serum albumin (BSA) blockade prior to immunolabeling. Primary immunoprobes—Anti-Myogenin (ab124800, Abcam, UK) and Anti-MHC (ab37484, Abcam)—were reconstituted in 1:200 diluent and applied for nocturnal (16 h, 4 °C) antigenic engagement. Following PBS washes, Alexa Fluor-secondary conjugates (1:400, Invitrogen) facilitated target visualization during 2-h ambient incubations. Nuclear architecture was demarcated through 10-min chromatin labeling with DAPI (1:1000).

4.11. Western blot analysis

After identical treatment durations and culture conditions, total cellular proteins from L6 cells were isolated using RIPA lysis buffer. Protein expression of MyoD, MyoG, and MHC was evaluated at specific time points following LIPUS stimulation: MyoD, an early-stage myogenic marker, was collected on day 4, while MyoG and MHC, late-stage markers, were collected on day 8 [54]. Protein samples were resolved via 10 % SDS-PAGE and subsequently electrophoretically transferred onto nitrocellulose membranes (Beyotime). The membranes were probed overnight at 4 °C with specific primary antibodies (details provided in Supplementary Table S2). Following incubation with HRP-linked secondary antibodies (1:1000, Invitrogen), an Odyssey infrared imaging system was used to detect immunoreactive bands. Densitometric analysis of protein expression levels was conducted with ImageJ software (NIH).

4.12. Transcriptome sequencing

Total RNA isolation was performed with TRIzol reagent (Invitrogen) following the manufacturer's guidelines. Polyadenylated mRNA was enriched via oligo(dT)-coupled magnetic beads (Dynabeads Oligo(dT)), followed by fragmentation in a magnesium-containing buffer at 94 °C for 5–7 min. First-strand cDNA synthesis was executed using SuperScript™ II Reverse Transcriptase, and second-strand cDNA was generated with E. coli DNA polymerase I and RNase H in the presence of dUTP to preserve strand orientation. The resulting cDNA underwent end-repair, adenylation, and adapter ligation (T-tailed adapters). Size selection targeting ∼300 bp (±50 bp) inserts was achieved using magnetic beads, after which the second cDNA strand was enzymatically digested with uracil-DNA glycosylase (UDG, NEB). PCR amplification was carried out under the following thermal cycling conditions: initial denaturation at 95 °C (3 min), 8 cycles of 98 °C (15 s), 60 °C (15 s), and 72 °C (30 s), with a terminal extension at 72 °C (5 min). Final libraries were quality-assessed and sequenced on the Illumina NovaSeq™ 6000 platform (LC Bio Technology Co., Ltd., China) to produce paired-end 150 bp (PE150) reads.

Raw sequencing data (FASTQ format) underwent preprocessing with fastp to eliminate adapter sequences, low-quality reads, and PCR duplicates. Cleaned reads were mapped to the Rattus norvegicus reference genome (assembly RGSC Rnor_6.0) via HISAT2. Aligned reads in BAM format were analyzed for transcript abundance using StringTie, with gene expression quantified as FPKM (Fragments Per Kilobase of Exon per Million Mapped Fragments). Differentially expressed genes (DEGs) were identified through edgeR, applying thresholds of |log₂ fold change| > 1 (equivalent to fold change >2 or < 0.5) and adjusted p-value <0.05. Functional annotation of DEGs was conducted through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses using the DAVID bioinformatics platform.

4.13. Role of intracellular Ca²⁺ concentration in myogenic differentiation

Verapamil (100 μM, Solarbio, China) was used as a Ca²⁺ channel blocker in the culture medium. After 4 days of incubation in myogenic medium with or without verapamil, intracellular Ca²⁺ concentration was quantified using a calcium colorimetric assay kit (Beyotime), with absorbance measured at 575 nm using a microplate reader. For fluorescence-based detection of intracellular Ca²⁺, Rhod-2/AM (4 μM, Yeason, China) was employed as the Ca²⁺ indicator. Specifically, the samples were washed with PBS and incubated with the Rhod-2/AM working solution at 37 °C for 30 min. After incubation, the samples were then visualized under a fluorescence microscope. To explore the potential mechanisms, immunofluorescence staining was performed using anti-NFAT2 (1:200, A1539, ABclonal, China) and anti-CACNA1S (1:200, A23062, ABclonal) antibodies, as described above. Additionally, Western blot analysis was conducted to investigate proteins potentially involved in the mechanistic pathway, with the detailed protocol provided above and the corresponding antibodies listed in Table S2.

4.14. Animal study

All experimental protocols strictly adhered to the National Institutes of Health (NIH) guidelines for the ethical treatment of laboratory animals and received formal approval from the Institutional Animal Care and Use Committee of Fudan University (Approval ID: 2024-HDYY-001). A total of 40 adult male rats (8–10 weeks old, approximately 300 g) were purchased and randomly assigned to four groups: the normal group (uninjured abdominal wall), the PCL group, the PCP group, and the PCP + LIPUS group. Pentobarbital sodium (30 mg/kg) was administered intraperitoneally to induce anesthesia. Following depilation of the abdominal region and sterilization with 75 % ethanol solution, a 4-cm longitudinal incision was created 2 cm lateral to the abdominal midline. A 2 × 1 cm² full-thickness abdominal wall segment (encompassing fascia, muscular layers, and peritoneum) was excised. A 3 × 2 cm² patch (PCL, PCP) was surgically implanted using an inlay technique, ensuring ≥0.5 cm peripheral overlap with the defect margins. The abdominal muscle layer and tissues were sutured with 4-0 silk sutures, and the skin was closed using interrupted 3-0 sutures. In addition, the PCP + LIPUS group received 10 min of daily LIPUS stimulation with the following parameters: frequency of 1 MHz, power density of 3 W/cm², and duty cycle of 20 % [53]. This relatively higher in vivo intensity was selected to compensate for LIPUS attenuation by skin, subcutaneous fat, and connective tissues, ensuring effective acoustic stimulation at the defect site [55].

4.15. Histological analysis

Upon completion of the in vivo study, the rats were sacrificed under ethical protocols at postoperative weeks 4 and 8. Surgical sites were re-excised by incising the healed cutaneous layer to assess defect regeneration. Explanted specimens were imaged for macroscopic evaluation, then immersed in 10 % neutral buffered formalin for 48 h. Following paraffin embedding, serial 6 μm coronal tissue slices were prepared using a Zeiss microtome. Deparaffinized sections underwent graded ethanol rehydration and were histochemically stained with H & E and Masson's trichrome per standardized protocols to analyze muscular architecture and extracellular matrix remodeling. Whole-slide digitalization was performed using a Pannoramic MIDI II scanner (3DHISTECH, Hungary).

For antigen visualization, deparaffinized sections underwent epitope retrieval and blocking prior to incubation with primary antibodies targeting MHC (1:200, ab37484, Abcam), CCN2 (1:200, A11067, Abclonal), and CACNA1S (1:200, A23062, ABclonal) at 4 °C for 16 h. Following PBS-T washes, Alexa Fluor-tagged species-matched secondary antibodies (1:400, Invitrogen) were applied for 60 min at ambient temperature. Nuclear labeling was achieved via DAPI (1:1000, Invitrogen) incubation for 10 min. Fluorescent signals were captured using a fluorescence microscope, with quantitative intensity analysis executed in ImageJ (NIH) using background-subtracted regions of interest.

4.16. Biomechanical testing

Biomechanical testing was performed using a universal testing machine (5982, Instron, USA). Briefly, the muscle specimens were securely clamped at both ends and subjected to uniaxial tensile loading until failure at a loading rate of 10 mm/min. Stress–strain curves were recorded, and Young's modulus and ultimate tensile strength were calculated accordingly.

4.17. Statistical analysis

All datasets (n = 3) were expressed as mean ± standard deviation and interrogated using GraphPad Prism 5 (GraphPad, Inc., USA). Intergroup variances were examined through unpaired t-tests (dual comparisons) or analysis of variance (ANOVA) with Tukey's post hoc correction (multigroup analyses). Statistical thresholds were defined at p < 0.05, with FDR adjustment applied for multiple hypothesis testing.

Ethics approval statement

All experimental protocols strictly adhered to the National Institutes of Health (NIH) guidelines for the ethical treatment of laboratory animals and received formal approval from the Institutional Animal Care and Use Committee of Fudan University (Approval ID: 2024-HDYY-001).

CRediT authorship contribution statement

Qian Wu: Investigation, Methodology, Visualization, Writing – original draft. Yisi Liu: Investigation, Methodology, Visualization. Heng Song: Investigation, Methodology, Resources, Validation. Zhicheng Song: Investigation, Methodology, Validation, Visualization. Hekai Shi: Methodology, Validation. Dongchao Yang: Data curation, Methodology, Validation. Xiaoyu Peng: Investigation, Methodology, Visualization. Binbin He: Investigation, Methodology, Resources. Junliang Ma: Data curation, Resources. Fengxuan Han: Conceptualization, Funding acquisition, Methodology, Writing – review & editing. Bin Li: Conceptualization, Funding acquisition, Resources, Writing – review & editing. Yan Gu: Conceptualization, Funding acquisition, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

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

Appendix A. Supplementary data

The following is the Supplementary data to this article.

Data availability

Data will be made available on request.