PLCL was chosen for electrospinning due to its good flexibility, biodegradability, and biocompatibility [35, 36]. PLCL/ZnO patch was fabricated via multi-channel electrospinning of a polymer mixture containing PLCL, ZnO nanoparticles, and I-2959 photoinitiator. Subsequently, PLCL/ZnO patch was immersed in MPC solution to grow a PMPC coating on each fiber of the top surface by UV curing, yielding JPLCL/ZnO patch. Bioinspired by the anisotropic architecture of human abdominal wall, JPLCL/ZnO patch with a biomimetic muscle fiber-like morphology was fabricated by adjusting fiber orientation during electrospinning (Fig. 2a). As shown in Fig. 2b and c, our patch exhibits a morphology with fiber alignment similar to that of muscle fibers, with fibers aligned at 0°, 45°, and 135°, corresponding to the transversus abdominis, internal oblique, and external oblique muscles, respectively. Moreover, our patch can be produced at a large scale (Fig. 2d), demonstrating its potential for clinical translation. These results demonstrate that our patch with a biomimetic muscle fiber-like morphology has been successfully fabricated.
In clinical practice, designing patches that match the mechanical anisotropy of natural abdominal wall is crucial for defect repair, as mechanical mismatch can cause poor tissue integration, postoperative adhesion, and hernia recurrence [31, 34, 37]. Typically, patches used for the repair of abdominal wall defect in various regions are expected to possess region-specific anisotropy ratios (the mechanical property ratio in the coronal axis and vertical axis). To evaluate the mechanical anisotropy of our patches, stress-strain curves of JPLCL and JPLCL/ZnO patches were analyzed under tensile loading applied either parallel or perpendicular to the fiber orientation (Figure S1). Both patches exhibit significant anisotropy, with the parallel direction exhibiting significantly higher tensile strength and elastic modulus than the perpendicular direction (Figure S2). Specifically, compared with JPLCL patch, JPLCL/ZnO patch demonstrates a 61% increase in tensile strength (7.71 vs. 4.79 MPa) and 174% increase in elastic modulus (7.14 vs. 2.60 MPa) in the parallel direction (Figure S2). The enhanced mechanical properties are attributed to the incorporation of ZnO nanoparticles, which act as rigid nanofillers in the PLCL polymer matrix to improve load transfer, restrict polymer chain mobility, and thereby increase tensile strength [38,39,40]. To meet the mechanical anisotropy requirements of diverse regions of the abdominal wall, the mechanical properties of our JPLCL/ZnO patches can be changed according to the actual application requirements by adjusting the orientation stacking of the fibers. By adjusting the volume ratios of the polymer mixture electrospun in the 0° and 90° directions from 1:1 to 10:1 (Fig. 2e and Table S1), JPLCL/ZnO patches with different orientation stacking ratios (i.e., 10:1, 5:1, 3:1, and 1:1) were fabricated and their mechanical properties were investigated. Our JPLCL/ZnO patches can achieve anisotropy ratios ranging from approximately 1 to 14 (Fig. 2f and Figure S3), which can meet the full spectrum of abdominal wall anisotropy requirements (a range of 1–9) [29, 30]. Moreover, the tensile strength (4.39–7.71 MPa) and elastic modulus (3.49–7.14 MPa) of these patches meet or exceed the baseline requirements (0.08 MPa for tensile strength [31, 32] and 0.04 MPa for elastic modulus [41]) for abdominal wall defect repair (Fig. 2f). In addition to fiber orientation, ZnO content influences the mechanical performance of JPLCL/ZnO patches. As shown in Figure S4, JPLCL/ZnO patches with higher ZnO contents exhibit increased tensile strength and elastic modulus, indicating the reinforcing role of ZnO nanoparticles in the PLCL matrix. Considering the anisotropic gradient distribution characteristics of human abdominal wall, we have selected JPLCL/ZnO patch with 5 wt% ZnO and the highest anisotropy as the representative sample for our study to investigate the effects of structural anisotropy on mechanical performance and tissue regeneration. Under a 200 g load, JPLCL/ZnO patch exhibits greater elongation in the perpendicular direction than in the parallel direction (Fig. 2g), demonstrating its high flexibility and mechanical anisotropy. Therefore, by adjusting the orientation stacking of the fibers, JPLCL/ZnO patches possess tunable anisotropy ratios, tensile strength, and elastic modulus, allowing them to adaptively match the mechanical requirements of natural abdominal wall.
The fiber alignment of porcine abdominal wall muscle was evaluated through optical microscope. As shown in Figure S5, the porcine abdominal wall exhibits an aligned fibrous architecture with visible fiber bundles. To further assess fiber morphology, tissue samples were harvested and sectioned in two directions for HE staining: longitudinal section (parallel to the muscle fiber direction) and cross section (perpendicular to the fiber direction). As shown in Figure S6, the muscle fibers are highly aligned with an average diameter of approximately 50 μm, which is consistent with reported diameters of porcine muscle fibers with a range of 47–98 μm [42, 43]. Inspired by the anisotropic architecture of the abdominal wall, we have fabricated JPLCL and JPLCL/ZnO patches to mimic its native structure for defect repair. The morphologies of these patches were characterized by scanning electron microscopy (SEM). JPLCL and JPLCL/ZnO patches exhibit fibrous morphologies, with an increased fiber diameter on the top surface (PMPC-coated surface) compared to the bottom surface following the deposition of the PMPC coating (Fig. 2h and i). Orientation analysis further reveals that most fibers are aligned within a narrow deviation of −20° to 20° (Figures S7 and S8), indicating the highly oriented fibrous structure. The fiber orientation observed in our JPLCL/ZnO patch closely mimics the highly aligned structure of porcine abdominal wall muscle, which is crucial for mimicking the anisotropic characteristics of natural abdominal wall in the application of defect repair.
The Janus structure of our JPLCL/ZnO patches was characterized using X-ray photoelectron spectroscopy (XPS) and water contact angle measurements. XPS spectra confirm the presence of P and N elements on the top surface of JPLCL/ZnO patch (Figure S9), indicating the successful formation of PMPC. As shown in Fig. 2j and k, the water contact angle is almost 0° on the top surface and 122° on the bottom surface of JPLCL/ZnO patch, indicating the superhydrophilic nature of PMPC coating. This highly hydrated surface can effectively resist protein adsorption and cellular attachment, thereby preventing tissue adhesion in defect repair [44, 45]. Moreover, JPLCL/ZnO patch exhibits a fibrous morphology with a rough surface, as indicated by the C, O, and P elements predominantly distributed along the fibers (Figure S10), indicating that PMPC coating does not disrupt the original fibrous morphology of JPLCL/ZnO patch. In contrast, we prepared a PLCL/ZnO patch with Janus structure using traditional benzophenone treatment (BPLCL/ZnO). It can be clearly seen from the SEM image (Figure S11) that the original fibrous morphology of BPLCL/ZnO patches is covered by a continuous PMPC coating, with C, O, and P elements uniformly distributed across the entire surface. Traditional benzophenone treatment could destroy the highly oriented morphology and mechanical anisotropy of the patches [44], leading to a mechanical mismatch with the abdominal wall and hindering the repair process. To better visualize the spatial relationship between fibers and coatings, JPLCL/ZnO patch was stained with Nile red for fiber skeleton and sodium fluorescein for PMPC coating, and subsequently observed under a fluorescence microscope. Red-stained fiber skeleton and green-stained coating are clearly visible on the top surface of JPLCL/ZnO patch, with the green-stained coating seamlessly conforming to the red-stained fiber skeleton (Fig. 3a), demonstrating that PMPC coating is formed on each fiber of the top surface. In contrast, only the red-stained fiber skeleton is observed on the bottom surface (Fig. 3b), further confirming the Janus structure of JPLCL/ZnO patch. As a conceptual characterization, we conducted a slope slide experiment by sliding two ceramic blocks over both surfaces of JPLCL/ZnO patch. The top surface of JPLCL/ZnO patch was stained blue with a 0.01% (w/v) erioglaucine disodium salt solution (Brilliant Blue FCF, an FDA-approved biocompatible dye [46, 47]) to distinguish it from the white bottom surface. As shown in Fig. 3c, the red ceramic block moves smoothly on the top surface, while no visible movement of the blue ceramic block is observed on the bottom surface (Video S1), indicating the excellent lubricating properties of PMPC coating. These results demonstrate that JPLCL/ZnO patch with a Janus structure has been successfully fabricated by constructing PMPC coating on each fiber on its top surface.
a) Schematic diagram of muscle fibers alignment in human abdominal wall: (i) The transversus abdominis (upper layer) exhibits muscle fibers aligned at 0°, while the internal oblique (lower layer) exhibits muscle fibers aligned at 45°. (ii) The internal oblique (upper layer) exhibits muscle fibers aligned at 45°, while the external oblique (lower layer) exhibits muscle fibers aligned at 135°. b, c) SEM images of a(i) region biomimetic JPLCL/ZnO patch (b) and a(ii) region biomimetic JPLCL/ZnO patch (c). d) Digital photo of large-scale JPLCL/ZnO patch. e) Schematic illustration of JPLCL/ZnO patches with different orientation stacking ratios. f) Quantitative analysis of anisotropy ratio, tensile strength, and elastic modulus of JPLCL/ZnO patches with different orientation stacking ratios. g) Digital photos of JPLCL/ZnO patch loaded with a 200 g weight in the parallel and perpendicular directions. h, i) SEM images of the top (h) and bottom (i) surfaces of JPLCL/ZnO patch. j, k) Water contact angles of the top (j) and bottom (k) surfaces of JPLCL/ZnO patch
Wireless electrostimulation therapy, such as ultrasound therapy, has shown great potential in promoting tissue repair due to its non-invasive nature and precise targeting capability [48, 49]. Piezoelectric materials can be wirelessly activated by ultrasound to produce localized electrical stimulation to promote cell behavior and tissue regeneration [50, 51]. The electric performances of JPLCL/ZnO patches with different ZnO contents were evaluated under ultrasound stimulation at a power intensity of 0.5 W cm−2 and a pulse duty of 50%. As shown in Figure S12, the output voltage of JPLCL/ZnO patch increases progressively with ZnO content. In contrast to JPLCL patch, JPLCL/ZnO patch exhibits an ~ 2.4-fold increase in output voltage (Fig. 3d), demonstrating the enhanced electrical performance with the incorporation of ZnO nanoparticles. Under ultrasound stimulation, our patch undergoes cyclic deformation that induces electrical polarization and produces localized electric signals, which have been shown to regulate cellular behaviors such as cell migration and proliferation [52,53,54]. Considering that the thickness of human abdominal wall typically ranges from 10 to 30 mm [55], an ex vivo implantation model was designed to simulate its clinical application. As shown in Fig. 3e, JPLCL/ZnO patch was encapsulated between two copper (Cu) electrodes and Ecoflex films to construct a flexible device, which was then implanted at different depths of porcine tissue. An ultrasound probe was used to deliver ultrasound stimulation through the porcine skin, and the output voltages of JPLCL/ZnO patch were measured and recorded using an oscilloscope (Fig. 3f). As the implantation depth in porcine tissue increases from 10 mm to 30 mm, the output voltages of JPLCL/ZnO patch show a slight decrease (Fig. 3g and Figure S13). Moreover, our JPLCL/ZnO patch maintains stable electrical output even at a depth of 30 mm, with no significant voltage differences between the initial and final cycles over 105 cycles (Fig. 3h), indicating that ultrasound therapy can effectively penetrate tissues and precisely focus on targeted areas in a non-invasive manner. Therefore, our JPLCL/ZnO patch can be wirelessly activated by ultrasound to generate localized electrical stimulation for defect repair.
a, b) Fluorescence microscopy images of the top (a) and bottom (b) surfaces of JPLCL/ZnO patch. c) Digital photos of the slope slide experiment by sliding two ceramic blocks on the top (blue color) and bottom (white color) surfaces of JPLCL/ZnO patch. d) Output voltages of JPLCL and JPLCL/ZnO patches under ultrasound stimulation. e, f) Schematic illustration (e) and digital photo (f) of voltage generation of JPLCL/ZnO patch in porcine tissue under ultrasound stimulation. g) Output voltages of JPLCL/ZnO patch implanted at different depths of porcine tissue. h) Cyclic stability of the output voltage of JPLCL/ZnO patch implanted at a 30 mm depth of porcine tissue during 105 cycles
Good biocompatibility and antibacterial properties are essential for implanted materials in tissue repair to ensure both biosafety and prevent infection [56,57,58,59]. The biocompatibility of JPLCL/ZnO patch was evaluated using live/dead staining and cell counting kit-8 (CCK-8) assays. The morphologies and cell density of the JPLCL/ZnO group are similar to those of the control group (Figure S14a). In addition, the quantitative analysis of CCK-8 assay shows no significant statistical difference in cell proliferation between the two groups (Figure S14b), demonstrating the good biocompatibility of JPLCL/ZnO patch. In vivo biocompatibility was further evaluated by implanting PCO and JPLCL/ZnO patches subcutaneously in the dorsal region of rats for 5 days, after which tissue samples were collected for HE and immunohistochemical staining, including CD68 (a macrophage marker) and IL-6 (an inflammatory factor). The JPLCL/ZnO group shows a lower inflammatory response compared to the PCO group, and the expressions of CD68 and IL-6 in the JPLCL/ZnO group are significantly lower than those in the PCO group (Figure S15), indicating that JPLCL/ZnO patch does not induce obvious inflammatory reaction. To further evaluate the in vivo biosafety of the patches under ultrasound stimulation, major organs (e.g., heart, liver, spleen, lung, and kidney) were collected for HE staining after rats were sacrificed. As shown in Figure S16, no evident tissue damage is observed in the organs, indicating that our patch under ultrasound therapy exhibits good biosafety in vivo. In addition, the antibacterial properties of JPLCL/ZnO patch against E. coli and S. aureus were evaluated using the agar plate incubation method. The JPLCL/ZnO+US group shows fewer colony-forming units (CFU) of E. coli and S. aureus compared to the control and JPLCL/ZnO groups (Figure S17a). JPLCL/ZnO patch shows bacteriostatic rates of 88% and 86% against E. coli and S. aureus, respectively, which are further enhanced to 94% and 93% under ultrasound stimulation (Figures S17b and S17c), suggesting a synergistic antibacterial effect between the patch and ultrasound therapy. These results demonstrate that JPLCL/ZnO patch exhibits good biocompatibility and antibacterial properties.
The migration and proliferation of fibroblasts are crucial indicators in the tissue repair process, as they produce various extracellular matrix components and cytokines to promote tissue regeneration [60,61,62,63]. To investigate the activity of fibroblasts cultured with the patch under ultrasound stimulation, cell migration and proliferation rates were evaluated by cell scratch test and cell proliferation assay. The in vitro cell scratch test results show that the JPLCL/ZnO+US group significantly promotes the migration of fibroblasts compared to the control and JPLCL/ZnO groups (Fig. 4a and b), which can be attributed to efficient guiding effect of electric fields under ultrasound therapy. Cell proliferation rate was quantitatively evaluated using CCK-8 assay. L929 fibroblasts in the JPLCL/ZnO+US group exhibit a higher cell count than those in the control and JPLCL/ZnO groups (Fig. 4c), demonstrating its ability to promote cell proliferation. In addition to promoting cell migration and proliferation, abdominal wall defect repair materials also need asymmetric regulation of cells to achieve anti-adhesion performance. Therefore, the cell adhesion behaviors of JPLCL/ZnO patches were assessed to evaluate the effect of the Janus structure on biological adhesion. L929 fibroblasts were seeded on both the top and bottom surfaces of JPLCL/ZnO patch and cultured for one day to capture the fluorescence staining images. As shown in Fig. 4d, more L929 fibroblasts adhere to the bottom surface of JPLCL/ZnO patch and grow along the anisotropic fibers, while only few L929 fibroblasts are observed on the top surface, indicating that our JPLCL/ZnO patch can promote directional cell growth while preventing tissue adhesion. In contrast, a random JPLCL/ZnO patch was fabricated at a drum rotation speed of 100 rpm, and L929 fibroblasts cultured on these random fibers exhibit multidirectional growth and attachment due to the lack of anisotropic topology (Figure S18). The bioadhesion behavior of JPLCL/ZnO patch was further evaluated in vivo using a rat abdominal wall defect model. After 14 days of implantation, the bottom surface of JPLCL/ZnO patch adheres well to the defect tissue, while its top surface can effectively prevent visceral adhesion (Fig. 4e). These results demonstrate that our JPLCL/ZnO patch can not only promote cell migration and proliferation, but also prevent visceral adhesion.
Except for cell migration and proliferation, macrophage polarization plays a crucial role in tissue repair by regulating inflammation and promoting healing [64,65,66]. To investigate the effect of JPLCL/ZnO patch on immunoregulation, the expression of CD86 (a marker of M1 macrophages) and CD206 (a marker of M2 macrophages) were evaluated using flow cytometry. As shown in Fig. 4f and g, the JPLCL/ZnO+US group exhibits the highest M2-like/M1-like macrophage ratio among all the groups, indicating that JPLCL/ZnO patch under ultrasound therapy can enhance M2 macrophage polarization.
a) Cell migration images of fibroblasts for 0, 8, 16, and 24 h in the PCO, JPLCL/ZnO, and JPLCL/ZnO + US groups (scale bars: 200 μm). b) Quantitative analysis of cell migration (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; *** adjusted P < 0.001; error bars = SD; data are presented as mean ± SD). c) CCK-8 assay of L929 fibroblasts cultured after 1, 2, and 3 days in the PCO, JPLCL/ZnO, and JPLCL/ZnO + US groups (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; ** adjusted P < 0.01; *** adjusted P < 0.001; error bars = SD; data are presented as mean ± SD). d) Fluorescence images of L929 fibroblasts cultured on the top and bottom surfaces of JPLCL/ZnO patch for 1 day (scale bars: 40 μm). e) Digital photos of soft tissue adhering to JPLCL/ZnO patch in repairing a rat abdominal wall defect on the 14th day after surgery (scale bars: 1 cm). f) Quantification of CD206 and CD86 on RAW 264.7 cells by flow cytometry. g) Quantitative analysis of CD206/CD86 ratio (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; *** adjusted P < 0.001; error bars = SD; data are presented as mean ± SD)
To evaluate the in vivo pro-healing and anti-adhesion properties of JPLCL/ZnO patch, a rat abdominal wall defect model was established by creating a 15 mm circular defect in the abdominal wall using a circular punch. On the 14th day post-surgery, rats were sacrificed to observe defect repair and adhesion formation (Fig. 5a). Representative photos of wound healing and visceral adhesion are shown in Fig. 5b, including the surgically created defect and sutured patch on day 0, and the implanted patch in situ and the defect site after patch removal on day 14. JPLCL/ZnO patch does not cause severe visceral adhesion, and its anti-adhesion effect is comparable to that of clinical PCO patch. After removing the patches to expose the defects, the JPLCL/ZnO+US group exhibits markedly increased healing compared with the other groups (Fig. 5b). To evaluate the mechanical properties of JPLCL/ZnO patch in vivo, stress-strain curves were analyzed before implantation and 14 days after implantation. JPLCL/ZnO patch shows excellent mechanical anisotropy on the 14th day (Fig. 5c), which is beneficial for defect repair and long-term implantation [31, 67, 68]. In addition, an in vivo electrical performance experiment was conducted to investigate the feasibility of voltage generation from the patch under ultrasound stimulation in a rat model. After anesthetizing the rat with 3% pentobarbital sodium via intraperitoneal injection, JPLCL/ZnO patch was implanted in the abdominal wall defect and then exposed to ultrasound waves, the output voltage was recorded using an oscilloscope (Fig. 5d and Video S2). Under an ultrasound power intensity of 0.5 W cm−2, our JPLCL/ZnO patch can generate an output voltage of 231 mV (Fig. 5e), confirming the feasibility of delivering electric stimulation to the defect under ultrasound therapy.
To further evaluate the pro-healing properties of JPLCL/ZnO patches, tissue samples were collected from the rat abdominal wall defects on the 14th day after surgery for histological analysis. As shown in Figure S19, HE staining results indicate that both the control and US groups exhibit incomplete healing, indicating that ultrasound therapy alone hardly promotes wound healing. In contrast to the PCO and JPLCL/ZnO groups, the JPLCL/ZnO + US group shows reduced infiltration of inflammatory cells and denser collagen bundles (Figures S20 and S21). Quantitative analysis reveals that the collagen density in the JPLCL/ZnO+US group (77.4%) is significantly higher than the PCO (36.8%) and JPLCL/ZnO (44.7%) groups (Fig. 5f), indicating its excellent pro-healing property. Moreover, immunohistochemical staining results reveal that the expression of CD68 is significantly lower in both the JPLCL/ZnO and JPLCL/ZnO+US groups compared to the PCO group (Fig. 5g and Figure S22). Double immunofluorescence staining for CD31 and α-SMA was performed to evaluate angiogenesis. CD31 serves as a marker of endothelial cells and reflects new blood vessel formation, while α-SMA is a marker of vascular smooth muscle cells and indicates vessel maturation. The expression levels of CD31 and α-SMA in the JPLCL/ZnO+US group are significantly higher than those in the PCO and JPLCL/ZnO groups (Fig. 5h and Figure S23), indicating that JPLCL/ZnO patch can promote vascular proliferation under ultrasound stimulation. These results provide reliable histological evidence supporting the pro-healing ability of our JPLCL/ZnO patch under ultrasound therapy.
a) Schematic diagram of the repair of abdominal wall defect in a rat model. b) Digital photos of wound healing and visceral adhesions formation of abdominal wall defects in the PCO, JPLCL/ZnO, and JPLCL/ZnO + US groups. Day 0: surgically created defect (left) and sutured patch covering the defect (right). Day 14: implanted patch in situ (left) and defect site after patch removal (right) (scale bars: 1 cm). c) Mechanical properties of JPLCL/ZnO patches in the parallel and perpendicular directions before implantation and after 14 days of in vivo implantation. d) Digital photo of voltage generation from JPLCL/ZnO patch under ultrasound stimulation in a rat model. e) Output voltage of JPLCL/ZnO patch under ultrasound stimulation. f) Quantitative analysis of collagen density (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; ** adjusted P = 0.002; *** adjusted P < 0.001; error bars = SD; data are presented as mean ± SD). g) Quantitative analysis of CD68 (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; *** adjusted P < 0.001; error bars = SD; data are presented as mean ± SD). h) Quantitative analysis of CD31/α-SMA (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; PCO vs. JPLCL/ZnO + US * adjusted P = 0.014; JPLCL/ZnO vs. JPLCL/ZnO + US * adjusted P = 0.031; error bars = SD; data are presented as mean ± SD)
To simulate a clinically relevant abdominal wall defect in human, a porcine abdominal wall defect model was established by creating a 2 cm circular defect in the abdominal wall using an ultrasonic scalpel. These defects were subsequently repaired using patches, which were secured with 4-0 silk braided sutures (Fig. 6a and b). The experimental treatments were divided into three groups: the Mismatch group receiving a JPLCL/ZnO patch that misaligned with the natural abdominal wall anisotropy direction, the Mismatch+US group receiving a JPLCL/ZnO patch misaligned with the natural abdominal wall anisotropy direction and ultrasound therapy, and the Match+US group receiving a JPLCL/ZnO patch aligned with the natural abdominal wall anisotropy direction and ultrasound therapy. For the Mismatch+US and Match+US groups, ultrasound treatment was administered at 0.5 W cm−2 for 10 min per day and applied six days per week. After 28 days, the pigs were euthanized and tissue samples were harvested for histological analysis. HE and Masson staining results indicate that the Match+US group exhibits superior tissue alignment and increased collagen deposition compared to the Mismatch and Mismatch+US groups (Fig. 6c and d). This suggests that the biomechanical matching and ultrasound therapy can synergistically promote the repair of abdominal wall defects. Immunohistochemical and immunofluorescence staining further reveal that the Match + US group shows lower expression of CD68 and higher expression of CD31 and α-SMA compared to the Mismatch and Mismatch+US groups (Fig. 6c and e, and 6f). These results demonstrate that our JPLCL/ZnO patch can effectively promote tissue alignment, collagen deposition, and vascular proliferation under ultrasound therapy. Therefore, our JPLCL/ZnO patch not only activates electrical stimulation to promote tissue repair under ultrasound therapy, but also integrates excellent mechanical anisotropy and anti-adhesion properties for the efficient repair of abdominal wall defects.
a, b) Schematic diagram (a) and surgical process (b) of the repair of abdominal wall defect in a porcine model (scale bars: 2 cm). c) Images of HE staining, Masson staining, immunohistochemical staining, and immunofluorescence staining for the Mismatch, Mismatch+US, and Match+US groups (scale bars: 200 μm) d) Quantitative analysis of collagen density (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; ** adjusted P = 0.008; *** adjusted P < 0.001; error bars = SD; data are presented as mean ± SD). e) Quantitative analysis of CD68 (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; *** adjusted P < 0.001; error bars = SD; data are presented as mean ± SD). f) Quantitative analysis of CD31/α-SMA (n = 3 independent samples; ANOVA followed by Tukey’s multiple comparisons; ** adjusted P =0.003; error bars = SD; data are presented as mean ± SD)
