Microfluidic 3D printing polyhydroxyalkanoates-based bionic skin for wound healing

Biomimetic scaffolds with extracellular matrix (ECM)-mimicking structure have been widely investigated in wound healing applications, while insufficient mechanical strength and limited biological activity remain major challenges. Here, we present a microfluidic 3D printing biomimetic polyhydroxyalkanoates-based scaffold with excellent mechanical properties and hierarchical porous structures for enhanced wound healing. This scaffold is composed of poly(3-hydroxybutyrate-4-hydroxybutyrate) and polycaprolactone, endowing it with excellent tensile strength (2.99 MPa) and degradability (80% of weight loss within 7 d). The ECM-mimicking hierarchical porous structure allows bone marrow mesenchymal stem cells (BMSCs) and human umbilical vein endothelial cells (HUVECs) to proliferate and adhere on the scaffolds. Besides, anisotropic composite scaffolds loaded with BMSCs and HUVECs can significantly promote re-epithelization, collagen deposition and capillary formation in rat wound defects, indicating their satisfactory in vivo tissue regenerative activity. These results indicate the feasibility of polyhydroxyalkanoates-based biomimetic scaffolds for skin repair and regeneration, which also provide a promising therapeutic strategy in diverse tissue engineering applications.

Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

Introduction
The healing of skin wound is a common and challenging clinical problem with a huge financial burden to healthcare systems worldwide [1,2].Traditional wound dressings simply cover the wound surface and absorb exudates, providing limited protection for the wounds [3].Thus, great efforts have been made to develop new wound repair materials, including biological dressings [4][5][6], synthetic dressings [7][8][9], and tissue engineered dressings [10][11][12].Notably, tissue engineering scaffolds with extracellular matrix (ECM)-mimicking structures have attracted increasing attention for wound healing applications [13,14].However, the existing tissue engineering scaffolds are generally simple in structure and insufficient in mechanical properties.For example, most of the reported electrospinning tissue skin scaffolds had a disordered fiber structure, and their mechanical strengths only reached 50-100 kPa [15,16].In addition, these tissue engineering scaffolds usually tend to be biologically inactive as their strength increases, which may cause adverse reactions such as tissue allergies and inflammation [17,18].Therefore, it is highly desirable to develop a tissue engineering scaffold material with high mechanical strength and biological activity for wound healing.
In this paper, we present a microfluidic 3D printing biomimetic polyhydroxyalkanoates-based scaffold with the desired features for loading multiple cells, and can effectively promote wound healing, as schemed in figure 1. Microfluidic technology has emerged as an effective tool to generate uniform and continuous fibers with controllability and micron characteristics [19,20].In particular, the integration of 3D printing and microfluidic technologies provides a more effective choice for fabricating customized and highly ordered 3D-structured fibrous textiles [21][22][23].However, tissue engineering scaffolds fabricated with microfluidic 3D printing have been rarely reported for would healing [24].Besides, poly(3-hydroxybutyrate-4hydroxybutyrate) (P34HB) has received increasing attention in tissue engineering due to its intrinsic biocompatibility, appropriate degradability, and flexible mechanical properties [25,26].More importantly, 3-hydroxybutyric acid (3-HB), the degradation product of P34HB, is a basic metabolic energy substance of the human body, providing basic nutrients for cell growth and tissue regeneration [27].To the best of our knowledge, there are few researches on PHB-based skin biomimetic scaffolds, and its practical efficacy needs to be further investigated [28,29].Therefore, it is of great significance to explore a microfluidic 3D printing strategy for preparing the P34HBbased scaffold with biomimetic hierarchical porous structures and desirable mechanical properties for wound healing applications.
Herein, we employed a facile microfluidic 3D printing device to prepare highly ordered P34HB/polycaprolactone (PCL) scaffolds with ECM-mimicking porous structures.This polyhydroxyalkanoates-based biomimetic scaffold was composed of P34HB and PCL with excellent tensile strength (2.99 MPa) and degradability (80% of weight loss within 7 d).It was demonstrated that the hierarchical porous structure allowed bone marrow mesenchymal stem cells (BMSCs) and human umbilical vein endothelial cells (HUVECs) to proliferate and adhere on this scaffold, indicating that the scaffold had good cytocompatibility in vitro.In addition, the cellladen anisotropic composite scaffolds significantly promoted the re-epithelization, collagen deposition and capillary formation in rat skin wound defects.Therefore, we believed that such polyhydroxyalkanoates-based scaffold with biomimetic porous structures and excellent mechanical properties is a promising candidate for skin wound healing, which provided new insights into the design of biomimetic scaffolds in cell-based therapy and diverse tissue engineering applications.

Structure characterizations
Firstly, polyhydroxyalkanoates-based microfibers were prepared using a custom-made capillary microfluidic chip (figure S1, support information available online at stacks.iop.org/MF/1/015401/mmedia)[30,31].Briefly, the P34HB/PCL blend solution was obtained by dissolving the P34HB and PCL powder in an organic DCM/DMF solvent and then pumped into the microfluidic chip.The P34HB/PCL microfibers were then generated with fast diffusion of DCM/DMF and rapid solidification of P34HB/PCL in an ethanol bath.Besides, adding sodium chloride crystals to the blended solution could impart a porous structure to the microfibers.Bright-field optical (figure 2(a)(i)) and scanning electron microscope (SEM, figure 2(a)(ii) and (iii)) images showed that the microfibers exhibited a regular cylindrical morphology with a uniform diameter and smooth surfaces at both macro and micro levels.The cross-sectional structure of the microfibers was dense and impact, indicating that they had good toughness.The diameter of P34HB/PCL microfibers could be changed by adjusting the liquid flow rate, indicating that the preparation process was controllable (figure S2, support information) [32].In addition, the 3D printed PHA fibers reported in the previous literature usually require hightemperature heating to melt the PHA, which could easily damage the material properties and cause environmental pollution [33].Our P34HB microfibers were prepared under mild conditions and did not need high temperature or pressure.The

Future perspectives
The development of tissue engineering scaffolds with both mechanical strength and biological activity has always been pursued in tissue engineering technology.This inevitably requires optimization and innovation for scaffold materials and manufacturing processes.In this paper, in terms of scaffold material selection, polyhydroxyalkanoates is a promising biodegradable biopolymer with flexibly adjustable mechanical properties, thermal properties, and biocompatibility.However, the main challenge hindering its commercialization is the high cost of production.Technologies such as genetic engineering, enzymatic and metabolic engineering are needed to further reduce the fermentation cost and increase the industrial yield.On the other hand, the microfluidic-assisted printing method has been employed to assemble the continuous porous microfibers into uniform 3D scaffolds, which avoided the demand of high temperature or pressure in the traditional fused deposition modeling 3D printing process.The integration of microfluidics and 3D printing technologies provides a more effective choice for fabricating customized and highly ordered 3D-structured fibrous textiles.In addition, we aimed to fabricate a hierarchical porous scaffold with high cell loading capacity and mechanical strength for macroscopic wound healing, so we only designed a secondary hierarchical porous structure scaffold in micron-level.There is no doubt that such a microfluidic-assisted printing method could be a versatile and promising strategy for the fabrication of scaffolds with high resolution and delicate structures considering the flexibility of both microfluidic devices and 3D printing technology.Finally, living cells provide good bioactivity for polymer scaffolds, and the ECM-mimicking structural scaffolds provide adhesion space and mechanical support for the cells.With the advantage of multi-disciplinary intersection, the combination of various types of cells and cell derivatives with tissue engineering scaffolds can provide new strategies and directions for the development of biomedical engineering.organic solvent could be effectively removed during the fiber solidification process without affecting the natural environment.
Furthermore, hierarchical porous P34HB/PCL scaffolds with biomimetic structures were prepared by a microfluidic chip-modified 3D printer (movie S1, support information).The size and shape of the scaffold were pre-set by software.Subsequently, the microfluidic chip was used as a 3D printing head to fabricate the scaffold.P34HB/PCL scaffolds with designed shapes and sizes could be easily prepared by changing the software design (figure S3, support information).Because the diffusion rate of DCM and DMF in ethanol solution was slower than the weaving rate of the 3D scaffold, the curing of microfibers was incomplete when the 3D structure is formed, resulting in a tight connection between each fiber layer (figures 2(b) and S4, support information).The average fiber diameter measured by ImageJ software was 247.36 ± 7.03 µm (figure 2(c)).More importantly, the P34HB/PCL scaffold was designed with ECM-mimicking hierarchical porous structure to load multiple active cells for enhanced wound healing.The primary 3D interconnected large pores (378.86 ± 14.04 µm, figure 2(d)) could facilitate the cell penetration and ingrowth into the scaffolds under in vitro co-cultivation conditions, leading to higher cell loading capacity for subsequent in vivo wound healing applications.The secondary microporous structure (12.36 ± 3.53 µm, figure 2(e)) could promote the cell adhesion, expansion, and proliferation on the scaffold, because cells were more likely to adhere to scaffold structures with similar diameters (3-20 µm) and maintain a high proliferation rate as previously reported [34,35].The hierarchical porous structures of our scaffolds matched well with hierarchical ECM porous structures, which could not only offer a wide enough space for cell encapsulation and proliferation, but also provide mechanical support for loading cells into the wounds.Therefore, we described this hierarchical porous structure of the microfiber scaffold as an ECM-mimicking structure.In addition, the smaller size of the porous structure would be not favorable for cell penetration, so we only designed a secondary hierarchical porous structure scaffold in micron-level.There is no doubt that the microfluidic-assisted printing method could fabricate hierarchical porous scaffolds with nanoscale tertiary porous structures considering the flexibility of both microfluidic devices and 3D printing technology.
The polymerization reaction between P34HB and PCL was verified by Fourier transform infrared (FT-IR) spectroscopy.The absorption peaks located at 2850 cm −1 , 1451 cm −1 , and 1053 cm −1 could be assigned to the C-H vibrating, CH 2 bending, and C-O stretching of P34HB component in P34HB/PCL composite scaffolds, respectively (figure 2(f)) [36].Besides, the absorption peak at 1727 cm −1 belonged to the C=O bond, which gradually elevated with the increase of the PCL contents.There were no additional chemical bonds and functional groups present in the FI-TR, indicating the P34HB and PCL were physically blended during the microfluidic 3D printing process.Furthermore, the thermal stability of the P34HB/PCL scaffold was examined by thermogravimetric analysis (TGA).As shown in figure 2(g), the thermal decomposition of the composite scaffold showed two parts of decomposition accumulation: the 1st part was concentrated on the decomposition of P34HB at 300 • C-320 • C, and the 2nd part was concentrated on PCL at 400 • C-450 • C. In addition, compared with the P34HB/PCL non-porous scaffold, the thermal degradation curve of the porous scaffold had no significant change.The above results confirmed that the composite scaffold was composed of P34HB and PCL, and the porous structure would not affect the physical properties of the scaffold.
The degradation characteristics of the P34HB/PCL scaffold were verified by in vitro enzyme degradation experiments (figure 2(h)).It was found that the P34HB/PCL scaffold hardly degraded in the PBS solution.In contrast, the weight loss of scaffold increased significantly in the high-concentration lipase (100 U) solution, maintaining only 20% of the initial weight at day 7.This could be explained by the fact that P34HB was a fatty acid ester polymer, the scaffold would be specifically degraded in the presence of lipase.Notably, the degradation rate of porous scaffold was faster than that of nonporous scaffold (81.2% vs 75.0%weight loss within 7 d).This could be attributed to that the porous structures in the scaffolds would increase the contact area between the enzyme and P34HB/PCL scaffold, and thereby accelerate the degradation of the scaffold.All above results indicated the good degradation performance of the porous P34HB/PCL scaffold.
In order to fabricate a cell-laden porous scaffold for wound healing, we needed to maximize the cell loading rate of the scaffold.Before the plasma treatment, the water contact angle of the P34HB/PCL porous scaffold was measured to be 111.3• ± 7.3 • , slightly lower than the non-porous scaffold with 121.5 S5, support information).Due to the inherent hydrophobicity of P34HB and PCL materials, the porous modification had less effect on the hydrophilicity of the scaffold, which led to an unsatisfactory cell-loading capacity because few cells adhered to the hydrophobic scaffold surface in our in vitro experiment.Therefore, to further improve surface hydrophilicity, the scaffolds were plasmatreated with oxygen (O 2 ) and argon (Ar) for 5 min.Fortunately, more cells were infiltrated into the scaffold porous structures after the plasma treatment.Moreover, the superhydrophilic properties of the P34HB/PCL scaffolds allowed cells to adhere along the scaffold backbone with a high proliferation rate, which could be attributed to that the wettability and surface topography may regulate cell adhesion by cytophilic interactions and size-matching effects as by reported by Jiang et al in Nature Reviews Materials [37].In a previous study, micropatterns containing superhydrophilic and superhydrophobic regions were used to form high density cell microarrays [38,39].Different cell types preferentially adhered and grew to confluence on the superhydrophilic regions, because the trapped air in the superhydrophobic barriers could inhibit the cell proliferation and migration across these barriers [40].

Mechanical properties
As for wound healing applications, the porous scaffold should provide sufficient mechanical strength for cell ingrowth and ECM deposition during tissue regeneration process.In this study, the mechanical properties of P34HB/PCL microfibers and scaffolds were evaluated using a universal mechanical testing machine (figure 3(a)).The composite scaffold exhibited good flexibility to stretch, twist, fold and rapidly return to its initial shape after the external force was removed (figure 3(b), figure S6, and movie S2, support information) [41].Figure 3(c) showed that the tensile strengths of single P34HB/PCL non-porous and porous microfibers were 2.99 MPa and 1.25 MPa, and the elongation at break was 224.1% and 215.6%, respectively.After calculation, Young's modulus of the non-porous and porous microfibers was 180.76 MPa and 103.91 MPa, respectively (figure 3(d)).The good elasticity enabled the P34HB/PCL microfibers to withstand greater deformation pressure.In addition, the mechanical strength of the network scaffolds has been further investigated, and their tensile curves presented obvious zigzag waveforms as shown in figure 3(e).This indicated that the microfibers were gradually broken one after another during the stretching process, and the fracture of a single microfiber had little impact on the whole scaffold.Besides, the tensile strength of the P34HB/PCL composite scaffolds was significantly higher than that of the pure P34HB scaffolds (figure 3(f)).Taken together, the P34HB/PCL scaffold possessed good flexibility and tensile strength, which could satisfy the demands in mechanical properties for wound healing applications.

In vitro cytocompatibility
Good cell adhesion and proliferation were key factors for the application of cell-laden biomimetic scaffolds for wound healing.Therefore, BMSCs and HUVECs were cultured to explore the in vitro biological performance of the P34HB/PCL biomimetic porous scaffolds.It was observed that the HUVECs extended along the microfibers and closely adhered to the surface of the P34HB/PCL scaffold with a spindle-shaped appearance after 3 d of incubation (figure 4(a)).A highdensity homogeneous cell layer was formed on the surface of the P34HB/PCL scaffold, in remarkably contrast to the non-oriented cell morphology in the control group.Furthermore, live/dead and cell counting kit-8 (CCK-8) assays were performed to evaluate cell viability.Live/dead cell staining revealed that HUVECs and BMSCs were short-shuttle and long-spindle, respectively.The cell density gradually increased with the culture time, and most cells were green, with little red dead cells (figures 4(b) and S7, support information).As shown in figures 4(c) and (d), the hierarchical porous scaffold could promote the adhesion and proliferation of HUVECs as compared to the control group.However, there is no significance among the three groups in the proliferation of BMSCs.SEM observation further confirmed that the HUVECs and BMSCs were distributed in clusters on the scaffold surface with a contracted appearance (figure S8, support information).These results indicated that the P34HB/PCL scaffold could significantly promote cell adhesion and proliferation, and an anisotropic composite scaffold loaded with HUVECs and BMSCs was successfully prepared for further wound healing.

In vivo wound healing
The cell-laden P34HB/PCL biomimetic scaffold was implanted into rat skin wounds to evaluate its in vivo regenerative activity (figure S9, support information).After scaffold implantation, it was observed that the wounds covered by the cell-laden scaffolds healed faster than those of the control group, and less swelling and redness of the wounds were observed (figure 5(a)).The wound in the cell-laden scaffold group was almost completely closed, and the surface crust had disappeared on day 12.The quantitative wound area was 5.92%, and the effect was significantly better than that in the non-cell-laden group (15.85%) and the control group (14.19%) (figure 5(b)).In addition, the wound closure rate of the non-cell laden group was slightly lower than that of the control group in the early stage of wound healing (before day 6), but there was no significant difference.This might be due to the insufficient bioactivity of the pure polymer scaffold, which acted as a physical barrier that delayed wound healing.With the gradual degradation of P34HB/PCL scaffolds over time, the wound closure of the non-cell laden group was almost the same as that of the control group in the late stage of wound healing (after day 6).Masson and H&E staining were further performed to evaluate the collagen deposition and maturation of the regenerated tissues (figures 5(c) and 6(a) and (b)).As compared to the non-cell-laden scaffold and control groups, the cell-laden scaffold group exhibited the highest collagen deposition and the most orderly collagen arrangement reflected by the red stained muscle fibers and deep blue stained collagen fibers, indicating the enhanced ECM reconstruction and tissue remodeling ability by the cell-laden scaffolds.
The inflammation and angiogenesis of the wounds were further evaluated by immunohistochemical staining.The inflammation was alleviated by the P34HB/PCL scaffolds, as fewer yellow stained areas of IL-6 and TNF-α in the scaffold groups than that in the control group (figures 6(c) and (d)).CD31/α-SMA double immunofluorescence staining was conducted to assess the vascularity in the newly formed tissues.The result showed that the cell-laden scaffold group had the highest blood vessel density in the wound beds compared with other groups (figure 6(e)).This might be attributed to the loading cells and gradual release of 3-HB during the degradation of P34HB.3-HB blocked the activity of histone deacetylase, which improved the ability of cells to resist oxidative stress, and thereby promoted tissue regeneration and inhibiting inflammation [42,43].In addition, the scaffold loaded HUVECs and BMSCs provided essential components for capillary formation, which in turn accelerated skin wound healing [44].All the above results indicated that the seeded cells might be one of the main reasons to induce wound healing.It should be noted that the cell suspension alone could not stay on the wound surface for a long time, the fluid would be lost in a very short time, and the cells could not adhere and proliferate, or even survive.Therefore, the scaffold free cell group could be not a comparable control group in our study.Although our experiments did not set the scaffold free cell group separately, our results confirmed the significant enhanced wound healing effect of the seeded cells via the comparison between the cell-laden and non-cell-laden scaffold groups.

Conclusion
In summary, we had successfully prepared a biomimetic porous scaffold with an ECM-mimicking structure for skin wound healing.The microfiber scaffold prepared by combining microfluidics with 3D printing had good controllability and stability.The blending of P34HB and PCL significantly improved the mechanical and degradation properties of the scaffolds and provided functional support for tissue regeneration.The hierarchical porous structure facilitated the adhesion of HUVECs and BMSCs on the scaffold, so the P34HB/PCL scaffold could be used as a cell carrier for cell delivery.Furthermore, anisotropic composite scaffolds loaded with HUVECs and BMSCs could promote re-epithelization, collagen deposition and capillary formation in the rat skin wound model, indicating that the excellent wound healing ability of the bionic P34HB/PCL scaffold.These results showed that polyhydroxyalkanoates-based biomimetic scaffolds had promising applications in the skin regeneration field, and also provided a therapeutic strategy for other tissue regeneration fields.

Microfluidic spinning
The capillary microfluidic chip was composed of a piece of glass slide and a conical capillary glass tube.The tip of the capillary glass tube was firstly polished to a diameter of 300 µm under an optical microscope and then fixed on the glass slide with epoxy resin.To fabricate the continuous microfibers, 0.5 g of P34HB and PCL (w/w = 5/5) powder was dissolved in 2.5 ml of DMF/DCM (v/v = 5/5) and stirred uniformly to prepare a mixed solution.Pumped the mixed solution into the device via a syringe pump (LSP01-2 A).The stream was then flowed into the absolute ethanol bath for fast diffusion of DMF and DCM solution to induce solidification of the P34HB/PCL microfibers.The P34HB/PCL microfibers with different sizes could be generated by adjusting the orifice diameters and the solution flow rates.

Microfluidic 3D printing
The P34HB/PCL scaffolds were fabricated using a custommade 3D printer with microfluidic chips.The size and shape of the scaffold were designed by the 3D printed models before printing.The nozzle movement speed and the filling speed of the print model could be changed according to the speed of the syringe pump.In addition, the height of the printed scaffolds could be controlled by adjusting the printing time.In a typical experiment, the syringe pump was set at a speed of 1.5 ml h −1 .The printing time, moving speed, and filling rate of the customized 3D printer was set at 20 min, 5 mm s −1 , and 80%, respectively.

Characterization
Optical microscopic images of the P34HB/PCL microfibers and scaffolds were obtained with a dissecting microscope (Olympus BX51, Japan).After lyophilization, field emission SEM (FESEM, SU8010, Japan) was used to characterize the structure and morphology of the fiber scaffolds.ImageJ software was used to evaluate the pore diameter distribution from the SEM image.FT-IR spectroscopy was obtained by an FT-IR spectrometer (TENSOR II, Germany).TGA (TGA8000, USA) was used to analyze the thermal stability of the scaffolds.Water contact angles were obtained at ambient temperature by using a contact angle measuring instrument (JGW-360D, China).The enzymatic degradation experiments of non-porous and porous scaffolds (10 mm × 10 mm) were performed in PBS and lipase solution (Macklin, China).The overall degradation rate was calculated based on the daily mass loss of the scaffold during the degradation process.Besides, the universal mechanical testing machine (5944, USA) was used to assess the mechanical strength of the microfibers and scaffolds.

In vitro cytocompatibility
The HUVECs were incubated within endothelial cell medium (ECM, Science Cell Research Laboratories, SD, USA) which were mixed with 1% penicillin/streptomycin (P/S) and 5% fetal bovine serum (FBS) solution in the incubator (5% CO 2 , 37 • C).The BMSCs were incubated within minimum essential medium alpha (α-MEM, Gibco, USA) supplemented with 1% P/S and 10% FBS solution.All the fabricated scaffolds were sterilized by UV irradiation and 75% ethanol overnight.Before the cellular experiment, scaffolds were immersed in complete medium overnight for pre-culture.Cell proliferation and viability assays were performed to evaluate the cytotoxicity of co-culture with the scaffolds.Specifically, the BMSCs and HUVECs were seeded into the P34HB/PCL porous scaffold group, respectively.Cells seeded into the non-porous scaffold and culture without scaffold were used as control (n = 4).After co-cultivation for 1, 3, and 5 d, each well was incubated with 10% CCK-8 solution for 1 h.A microplate reader (Varioskan LUX, Thermo, USA) was used to measure the optical density values of the solutions.Live/dead cell staining of HUVECs and BMSCs on the scaffold was determined via the AM/PI staining kit.After culturing for 5 d, the scaffolds were stained with AM/PI for 30 min in the dark.Finally, an inverted fluorescence microscope (ZEISS Axio Vert.A1, Germany) was used to take the fluorescence images.
For the cell attachment experiment, P34HB/PCL nonporous and porous scaffolds were placed into a 24-well plate (n = 3).50 µl of BMSCs suspension (cell density of 10 5 ml −1 ) was first added to the scaffold surface.After the cells adhered for 4 h, 1 ml of medium was supplemented gently along the edge of the culture plate.Cell morphology observation was performed after 24 h, subsequently.The scaffolds were then fixed with 4% paraformaldehyde for 15 min.Lastly, scaffolds were incubated with F-actin for 30 min and stained with DAPI before the observation by a laser confocal microscope (Nikon A1, Tokyo, Japan).In addition, the above scaffolds were washed with PBS and dehydrated gradually in ethanol solution.After freeze-drying overnight, the morphology of the cells on the scaffold was observed by SEM.

In vivo wound healing
The cell-laden P34HB/PCL scaffolds were implanted into the skin wound of the rats to assess its in vivo wound healing performance.Briefly, eight male SD rats were anesthetized with pentobarbital, and three full-thickness circular wounds (12 mm in diameter) were made on the upper back of each rat (figure S9, support information).The customized anisotropic composite scaffolds loaded with HUVECs and BMSCs were implanted into the wounds and labeled as the cell-laden group.The wounds treated with non-cell laden scaffolds and without treatment were used as control groups.The periphery of the scaffolds was fixed with sterile sutures and covered with 3 M Tegaderm TM (Neuss, Germany).Then, the wound healing process was monitored and photographed every day.All the rats were sacrificed after 12 d, and the skin samples from the wound sites were harvested in a full layer in conjunction with surrounding tissues and then soaked in 10% formalin solution.Paraffin embedding was performed for subsequent histological analysis.H&E and Masson staining were used to observe tissue cell morphology and collagen deposition.Immunohistochemical staining was used to assess tissue inflammation and vascularization.

Statistical analysis
SPSS software (Version 22.0, USA) was used for statistical analysis.The statistical description of the data in the form of mean ± standard deviation (SD).A one-way analysis of variance and independent sample t-test methods were used to evaluate the significant differences.The significance level was marked with * p < 0.05, * * p < 0.01, * * * p < 0.001.

Figure 1 .
Figure 1.(a) Schematic diagram of the manufacturing process of the cell-laden P34HB/PCL biomimetic microfiber porous scaffold.(b) Schematic illustrations of wound healing using the cell-laden P34HB/PCL biomimetic scaffold.

Figure 3 .
Figure 3. Mechanical strength of P34HB/PCL microfibers and scaffolds.(a) Digital images of the tensile test of P34HB/PCL fibers and scaffolds using a universal mechanical testing machine.(b) Digital images showing the good flexibility of P34HB/PCL scaffolds.(c), (d) The strain-stress curves (c), elongation at break, and Young's modulus (d) of P34HB/PCL microfluidic fibers.(e), (f) The strain-stress curves (e), elongation at break and Young's modulus (f) of P34HB/PCL scaffolds.