Oriented Fibers Cooperate with DFO to Prevent Tendon Adhesions by Improving the Repair Microenvironment

To solve the problem of tendon adhesion after an operation, an anti‐adhesion membrane is designed, which can inhibit the exogenous healing of tendons and promote endogenous healing. Here, poly[3(S)‐methyl‐morpholine‐2,5‐dione‐co‐lactic acid] P(MMD‐co‐LA) containing alanine units is obtained by melt ring‐opening polymerization (ROP). It can effectively reduce the production of acidic degradation products that cause aseptic inflammation. Then a kind of oriented P(MMD‐co‐LA)/deferoxamine (DFO) anti‐adhesion membrane with good mechanical properties, cell adhesion properties, and induced macrophage polarization properties is constructed by electrospinning. In vitro and in vivo studies have shown that oriented fibrous membranes can down‐regulate the expression of inflammatory cytokines. In addition, the fibrous membrane can up‐regulate the expression of angiogenesis‐related genes, namely HIF1‐α, SDF‐1α, and VEGF, by releasing DFO in situ, thus promoting the revascularization of tendon defects and providing nutritional support for endogenous tendon healing. This method provides a new barrier strategy to reduce tendon adhesion through anti‐inflammation combined with vascularization to inhibit exogenous tendon healing while promoting tendon endogenous healing. The results provide essential insights into the corresponding regulation of the tendon healing microenvironment.

endogenous healing and inhibiting exogenous healing is critical to reducing tendon adhesion. [22,23] Exogenous healing mainly involves inflammatory cells outside the tendon or cells from adjacent tissues, such as paratendinous, and adventitia, such as fibroblasts. It invades the healing site from the tendon sheath and synovium around the tendon sheath, resulting in excessive deposition of the collagen matrix, and finally leading to scar formation and tendon adhesion with the surrounding tissue, which is also the main reason for tendon adhesion to the surrounding tissue. [22,23] In this regard, because of the small pore size of the electrospun fiber membrane, it can be used as a barrier to separate exogenous cells, such as fibroblasts, from tendon tissue in the tendon sheath. In addition, macrophages play an essential role in tendon remodeling after injury. They are highly malleable and can be activated by various stimuli of the surrounding environment. Macrophages can secrete inflammatory cytokines, remove fragments and form growth factors by activating inflammatory signals and cytokines through "classic" (M1) and "alternative" (M2) activation in the tendon healing process. Moreover, the M2 macrophages have apparent anti-inflammatory effects; thus, they will show some therapeutic potential. [24][25][26] Moreover, studies have shown that biomaterials can regulate the immune microenvironment by promoting the polarization of M2 macrophages. [27] Therefore, the oriented nanofibers regulate the immune microenvironment to prevent tendon adhesion by guiding macrophages to M2 phenotypic polarization, which may be an effective way to achieve immune-mediated prevention of tendon adhesion.
However, the simple oriented fiber membrane cannot meet the need for rapid promotion of angiogenesis. The strategy of loading exogenous angiogenic agents should be considered in preparing the fiber membrane because tendon cells and tissues need to provide timely nutrition in endogenous healing to promote the secretion of extracellular matrix and tendon healing. [28] Deferoxamine (DFO), an iron chelating agent approved by the FDA for transfusion iron overload, has been shown to promote angiogenesis and inhibit iron-induced hydroxyl radical formation by stabilizing HIF1-α and can improve tissue damage repair. [29,30] Among them, the rapid construction of angiogenesis in the injured site is essential in the tendon repair process, ensuring an adequate supply of nutrients in the tendon tissue repair. [23] In addition, oxidative stress is another factor in the progression of the disease. Inflammation and ischemia after tissue injury will accumulate reactive oxygen free radicals(ROS), producing oxidative damage to the tendon. [31] Therefore, controlling the oxidative stress in tendon injury is also important to prevent tendon adhesion. [3] DFO can form a stable complex with free iron, thus reducing the availability of free iron in ROS so that it can play a tremendous protective role. [32][33][34] Furthermore, tendon injury belongs to tissue injury; DFO has been used to promote the angiogenesis of skin tissue, peripheral nerve, and antioxidant stress; [29,35] thus it is expected to promote tendon endogenous healing angiogenesis and antioxidant stress injury. However, DFO promotes angiogenesis in a threshold-dependent manner. Long-term release cannot be achieved using DFO alone, and excessive release will produce different effects. [36] Therefore, to solve these problems, it is necessary to transport drugs locally. An electrospun fiber membrane can deliver antiadhesion drugs because of its unique characteristics. [37][38][39] In addition, in previous studies, we have verified that the long-term stable release of DFO can be achieved by compounding DFO into electrospun fibers. [35] Therefore, loading and releasing DFO from oriented fibers are promising to promote endogenous tendon healing and prevent tendon adhesion.
An appropriate microenvironment improves tendon adhesion and promotes tendon healing and functional recovery. However, it is affected by the lack of effective guidance and regulation or the interference of inflammatory response, ischemia, and excessive ROS. Here, we were committed to constructing composite scaffolds with oriented structure and drug reagents and introducing an oriented poly[3(S)-methyl-morpholine-2,5-dione-colactic]/DFO (P(MMD-co-LA)/DFO) fiber membrane to regulate the regenerative microenvironment ( Figure 1A). The regulation of macrophage polarization on random and oriented membranes was evaluated in vitro and in vivo ( Figure 1D). Furthermore, the effect of DFO-loaded fibrous membranes on enhanced human umbilical vein endothelial cells (HUVECs) angiogenesis was analyzed ( Figure 1C). In addition, we further verified the feasibility of P(MMD-co-LA) membrane loaded with DFO to inhibit tendon adhesion, promote tendon healing and functional recovery in rat tendon injury model ( Figure 1B,E), and found a new way to prevent tendon adhesion.

Synthesis of P(MMD-co-LA) Copolymers
P(MMD-co-LA) was obtained from the copolymerization of lactide (Aladdin) and MMD (homemade in the laboratory). Briefly, 3 g of lactide, 1 g of MMD, 2 mg of Sn(Oct) 2 (Sigma-Aldrich), and 2 mL of chloroform solution (Sinopharm Chemical Reagent Co., Ltd.) were added to a sealed reaction tube. The chloroform was evaporated under reduced pressure at low temperature and then dried under reduced pressure at 50 °C for 1 h using a mechanical pump. The reaction tube was vacuum sealed under an alcoholic torch and immersed in an oil bath for 16 h at 140 °C. Finally, the product was dissolved in dichloromethane (Sinopharm Chemical Reagents Co., Ltd.) and settled three times more than hexane (1:10, Sinopharm Chemical Reagents). The resulting P(MMD-co-LA) was dried under vacuum at room temperature for 24 h and stored at low temperature.

Preparation of Electrospun Membranes
The electrospun tendon anti-adhesion membrane was constructed by electrospinning technology. Briefly, a 20% (w/v) solution of P(MMD-co-LA) in Hexafluoroisopropanol (Aladdin) was prepared. Then 10 mg of DFO was added to 5 mL of the solution and stirred well to obtain the carrier mixture. The resulting spinning stock solution was drawn into a 10 mL syringe fitted with a 22-gauge needle. Electrostatic spinning was carried out using a high voltage of 9 kV (Beijing Yongkang), a solution flow rate of 0.6 mL h-1, a collection distance of 15 cm, and a collection drum speed of 2800 (oriented fibers) and 200 rpm (random fibers). The resulting fibrils were vacuum dried for 24 h and stored at −20 °C. The resulting random P(MMD-co-LA) fibrous membranes, oriented P(MMD-co-LA) fibrous membranes, and oriented . Schematic diagram of electrospun P(MMD-co-LA) oriented fiber membrane loaded with DFO and its biological function of reducing tendon adhesion and promoting tendon healing by anti-inflammation and vascularization in the tendon injury model. A) Schematic diagram of the preparation process of A_PDPLA/DFO electrospinning membrane. B) Schematic diagram of rat tendon injury model. C) a) DFO is released from oriented P(MMD-co-LA) fibers and chelated with Fe 2+ to prevent HIF1-α degradation; b) mechanism of DFO promoting tendon angiogenesis. D) Oriented fibers regulate macrophage polarization. E) The improved microenvironment promotes tendon repair and reduces tendon adhesion.

www.advmatinterfaces.de
To test the release rate of DFO, phosphate-buffered saline (PBS) was prepared with different concentrations of DFO. The solution was then mixed with an excess of FeCl 3 , and the absorbance of the mixed solution at 485 nm was measured to determine the DFO concentration versus absorbance. A_PDPLA/DFO was immersed in 10 mL of PBS and placed on a shaker at 37 °C (100 rpm). After a certain period, 0.9 mL of release medium was added to 0.1 mL of FeCl 3 (10 mg mL −1 ) solution. The corresponding absorbance was measured at 485 nm using an enzyme marker, and the amount of DFO released was calculated from the standard DFO calibration curve.

Cell Biocompatibility Evaluation
The electrospun fiber membrane was cut into a disc with a diameter of 30 mm by a punch, then sterilized with 75% ethanol for 2 h, irradiated under ultraviolet light for 24 h, then washed with PBS many times until the ethanol was removed entirely. First, tenocytes were inoculated into the 6-well plate at the density of 2 × 10 5 cells mL −1 (2.5 mL culture medium) per well; then the sterilized membrane was put into the well plate; after 24 h of co-culture in the incubator, the membrane was taken out, then the cells were digested with Trypsin-EDTA (Gibco, Canada), and the cells were collected by centrifugation (1000 rpm, 3 min). Then the supernatant was removed, and the PBS was added to prepare the cell suspension; the above steps were repeated three times. And 100 µL staining solution was mixed with 200 µL cell suspension and cultured at 37 °C for 15 min. Then the living cells with yellow-green fluorescence and the dead cells with red fluorescence were observed at the excitation wavelength of 490 ± 10 nm.

Polarization Experiment of Macrophages
The electrospun fiber membrane was cut into a disc with a diameter of 30 mm by a punch, then sterilized with 75% ethanol for 2 h, irradiated under ultraviolet light for 24 h, then washed with PBS many times until the ethanol was removed entirely. First, the sterilized membrane was fixed in a 6-well plate. Then RAW 264.7 macrophages were inoculated on the material at the density of 2 × 10 5 cells mL −1 (2.5 mL culture medium) per well. The high-glucose DMEM containing 10% FBS and 1% PS were added for 24 h and then stained with fluorescence. And first, the culture medium was sucked out, then the different fiber membranes were gently rinsed with PBS buffer three times, and the fiber membrane was fixed with 4% paraformaldehyde solution (Beyotime, China) for 10 min, and the fiber membrane was gently rinsed with PBS buffer for three times to remove the paraformaldehyde solution, then 0.5% Triton X-100/PBS (Beyotime, China) solution was added to permeate 5 min, and then 3% Bovine serum albumin (BSA, Beyotime, China) was added to incubate for 30 min. Then the CD206 (dilution 1/500, Abcam, USA) was added to the culture overnight at 4 °C and incubated with secondary antibodies for 50 min. And the TRITC Phalloidin (dilution 1/200, Solarbio, China) was added to incubate with the culture in the dark at 37 °C for 1 h. The DAPI was added to the stain at room temperature for 10 min. Finally, a positive fluorescence microscope (NIKON ECLIPSE C1, NIKON, Japan) was used to observe the immunofluorescence staining images of cells of the Control group and R_PDPLA, A_PDPLA and A_PDPLA/DFO membrane groups.

Wound Scratch Assay
HUVECs containing 2 × 10 5 cells mL −1 were inoculated into a 6-well plate. After 24 h of culture, the monolayer of cells was scraped with 10 µL gun head and added with A_PDPLA/DFO membrane extract (0.1 g membrane was extracted for 3 days in 10 mL medium). Then the cells were further cultured with the extract. The migration of HUVECs was photographed by an inverted fluorescence microscope (ICX41, SOPTOP, China) at 0, 12, and 24 h.

Endothelial Cell Tube Formation Experiment
To detect the effect of DFO on promoting vascularization, A_PDPLA/DFO membrane extract (0.1 g membrane was extracted for 3 days in 10 mL medium) was co-cultured with HUVECs in 6-well plate for 24 h, then the digested cells were prepared into three groups of cell suspensions. The HUVECs density was adjusted to 2 × 10 5 cells mL −1 . The 24-well plate and the gun head of 1 mL were pre-frozen in a refrigerator at 20 °C, and Matrigel (BD, USA) was thawed overnight at 4 °C. The Matrigel matrix glue was added to the 24-well plate at the ratio of 250 µL well −1 , then the orifice plate was placed in a cell incubator at 37 °C for 1 h to solidify, then 500 µL cell suspension was added to each hole, and the culture plate was placed in an incubator to form a tube. After being cultured for 3 and 6 h, the cell morphology and tube formation were observed under a light microscope (ICX41, SOPTOP, China), and the average tube formation value of the three pore cells was calculated by ImageJ v1.8.0 software.

Immunofluorescence Assay of Endothelial Cells
The electrospun fiber membrane was cut into a disc with a diameter of 30 mm by a punch, then sterilized with 75% www.advmatinterfaces.de ethanol for 2 h, irradiated under ultraviolet light for 24 h, then washed with PBS many times until the ethanol was removed entirely. HUVECs were inoculated into a 6-well plate at the density of 2 × 10 5 cells mL −1 (2.5 mL culture medium), and the A_PDPLA/DFO membrane extract was added (0.1 g membrane was extracted for 3 days in 10 mL medium). Then fluorescent staining was performed after 24 h of co-culture in the incubator. First, the culture medium was sucked out, the different orifice plates were rinsed gently with PBS buffer three times, and the cells were fixed with 4% paraformaldehyde solution for 10 min. Then the cell was gently rinsed with PBS buffer three times to remove the paraformaldehyde solution. Then the primary antibodies against HIF1-α (dilution 1/500, Bioss, USA), VEGF (dilution 1/400, Proteintech, China) and Cluster of differentiation 31 (CD31, dilution 1/1000, Proteintech, China) were added to the culture overnight at 4 °C and incubated with secondary antibodies for 50 min. And then, the DAPI was added to stain in the dark at room temperature for 10 min. Finally, a positive fluorescence microscope (NIKON ECLIPSE C1, NIKON, Japan) was used to observe the immunofluorescence staining images of cells of the Control group and R_PDPLA, A_PDPLA and A_PDPLA/DFO membrane groups.

Quantitative QPCR Assay of Macrophage Polarization and Angiogenesis
To evaluate the effects of A_PDPLA membrane on the expression of CD206, TNF-α, and IL-10 genes in RAW 264.7 macrophages, and the effects of A_PDPLA/DFO membrane loaded with DFO on the expression of HIF1-α, VEGF, and CD31 genes in HUVECs, real-time fluorescence quantitative polymerase chain reaction (RT-qPCR) was used. RAW 264.7 macrophages and HUVECs were inoculated on 6-well plates at 2 × 10 5 cells mL −1 density and co-cultured with oriented A_PDPLA membrane and extract for 3 days, respectively. According to the manufacturer's protocol, total RNA was extracted from mouse and human cells with TRIzol reagent (Invitrogen, USA) and ran on an RT-qPCR system (CFX Connect, BioRad, USA). The sequences of primer used for RT-qPCR analysis are shown in Table 1.

Animal Surgery
The rat tendon injury model was used to study the effect of electrospun fiber membrane on preventing tendon adhesion and promoting tendon healing. The Experimental Animal Welfare Ethics Committee of Zhongnan Hospital of Wuhan University had reviewed the animal experiment program. The use and care of animals were carried out under the animal experimental guidelines and regulations of Hubei Medical Experimental Animal Center. All electrospun fiber membranes were sterilized by Co60 irradiation before operation. The experimental animals were 80 SPF-grade 6-week-old male Sprague-Dawley (SD) rats (200-220 g). They were adapted to the SPF feeding room for 1 week before the operation. Eighty rats were randomly divided into four groups: control group, R_PDPLA, A_PDPLA, and A_PDPLA/DFO membrane groups. After the rats were anesthetized by intraperitoneal injection of pentobarbital sodium (40 mg kg −1 ), the prone position was tied to the fixed plate; the hair was cut off from the distal end of the hindlimb (including claws) with scissors, and then the skin was disinfected with iodophor. According to the aseptic requirement, an incision was made on the inside of the plantar surface (starting from the interphalangeal space between the first and second toes to the ankle), and the skin and subcutaneous tissue were separated in turn. The superficial flexor tendon of the toe was removed, and the tendon of the deep flexor digitorum profundus(FDP) was transected. The broken end of the tendon was sutured with a 6-0 Prolene suture (Kessler method). The skin of the control group was sutured directly after washing, and the other experimental groups were washed and sutured after wrapping the corresponding electrospun fiber membrane. The experimental animals were rewarded with a baking lamp until they woke up, and the animals returned to the cage position after the operation. [40]

Macroscopic Evaluation
To evaluate the severity of adhesion around the tendon, the adhesion around the tendon was graded according to Yang and other [41] grading criteria: grade 5, adhesive tissue with an area of more than 97.5% needed sharp separation; grade 4, adhesive tissue with an area of 51%-97.5% needed sharp separation; grade 3, adhesive tissue with an area less than 50% needed sharp separation; grade 2, adhesive tissue could be separated by blunt separation alone; grade 1, no adhesion was formed. The adhesion rate was used to quantify the degree of tendon adhesion. Two independent observers assessed the degree of adhesion by a blind method. [42]

Biomechanical Evaluation
The maximum tensile strength of the tendon was measured by an electronic universal tensile test machine (E44.104, MST industry, China) to evaluate the adhesion and healing of the tendon. First, the proximal and distal ends of the tendon were fixed to the dynamometer, and then the brake pulled the tendon at the speed of 10 mm min −1 until the end of the tendon broke, in which case the rheometer recorded the maximum tension. [43] Adv. Mater. Interfaces 2023, 10, 2300054

Histological Evaluation
The samples were taken 3 weeks after the operation, and the histology of the tendon was analyzed. The collected tendon tissues were fixed in 4% paraformaldehyde solution for 24 h and then embedded in long strips in paraffin to prepare 3 µm thick sections, which were stained with hematoxylin, eosin (HE) and Masson. Tendon adhesion and tendon healing were evaluated histologically. The histological degree of tendon adhesion was divided into four grades: grade 1, no adhesion; grade 2, mild (tendon surface adhesion less than 33%); grade 3, moderate (tendon surface adhesion between 33% and 66%); and grade 4, tendon surface adhesion greater than 66%. [44] Tendon healing was graded according to Tang and other grading criteria: grade 1, excellent (continuous collagen fibers in the tendon, smooth around the tendon); grade 2, good (collagen fibers in the tendon showed good repair, but the tendon sheath was invaded by adhesive tissue); grade 3, general (irregular collagen bundles in the tendon, partly invaded by adhesive tissue); grade 4, poor (partial separation or proliferation of much granulation tissue at the suture). [45] Two independent observers evaluated the histological degree by a blind method under an optical microscope (NIKON ECLIPSE C1, NIKON, Japan). [42]

Western Blot Analysis
To verify the effect of an electrospun fiber membrane loaded with DFO on preventing tendon adhesion, the expressions of the adhesion markers collagen I and collagen III, were detected by Western blot. The adhesive tissues were collected and put into a 1.5 mL centrifuge tube. 200 µL RIPA (Beyotime, China) lysate, 2 µL PMSF (Thermo, USA) and 2 µL protease inhibitor (Thermo, USA) were added to the tissue homogenizer. The tissues were homogenized for 2 min until the tissues were fully broken, then they were placed on ice for 30 min and centrifuged at 12 000 rpm at 4 °C for 10 min and stored on ice in a new 1.5 mL centrifuge tube. BCA protein concentration determination kits (TANGEN, China) were used to determine the total protein concentrations of different samples. The sample was electrophoretic by 10% SDS-PAGE gel and then transferred to a PVDF membrane (Millipore, USA). After sealing with 5% skim milk, the membrane was placed overnight at 4 °C with anti-collagen III (dilution 1/1000, Proteintech, China) and collagen I (dilution 1/1000, Proteintech, China) and GAPDH (dilution 1/5000, Proteintech, China). After washing, the membrane was incubated with secondary antibodies (Proteintech, China) for 1 h. The membrane was washed three times with TBST buffer (50 mm Tris-HCl, 100 mm NaCl, and 0.1% Tween-20, pH 7.4) and scanned with an imaging system (Image Quant LAS 4000 mini, GE).

Immunofluorescence Assay
The phenotype of macrophages, the expression of related inflammatory factors, and angiogenesis was detected by immunofluorescence staining on tendon paraffin sections. After dewaxing and dehydration, the slices were sealed with BSA as described above. Some of the slices were incubated overnight

Quantitative qPCR Assay
The tendon tissues of the control group and the experimental group were collected 3 weeks after the operation, and the expressions of inflammatory indexes TNF-α and IL-10, oxidative stress indexes MnSOD, HO-1, and vascularization indexes HIF1-α, SDF-1α, VEGF, and CD31 were detected by RT-qPCR.
According to the manufacturer's protocol, total RNA was extracted from tendon tissue with TRIzol reagent (Invitrogen, USA) and ran on an RT-qPCR system (CFX Connect, BioRad, USA). The primer sequences used in RT-qPCR analysis are shown in Table 2.

Statistical Analysis
The  were used for statistical analysis. And the statistics of multiple comparisons were carried out by single factor analysis of variance or t-test. All quantification data were presented as mean ± standard deviation (SD). The threshold of p < 0.05 was used to determine the statistical significance. Figure 2A shows the mechanism of co-polymerization of LA and MMD into P(MMD-co-LA). Figure 2B a) 1 H NMR results show that the MMD has a high purity and can be used for subsequent polymerization. The SEM results show that the electrospun fibers are continuous and circular with few beads ( Figure 2C). The results of the mechanical properties of the fibrous films show that the ultimate tensile strength (TS) and Young's modulus of the oriented fiber orientation direction (A + _PDPLA) are significantly higher than those of the randomly oriented fiber (R_PDPLA). The oriented fibers (A − _PDPLA) can withstand greater deformation in the vertical direction, showing considerable reliability ( Figure 2D). The UV absorption spectroscopy results show that DFO can be consistently released from the fiber membrane indicating that DFO has been successfully loaded into the fibers. We find an explosive release of DFO (≈30%-40%) from the solution within 1 h. This may be related to the fact that DFO is a hydrophilic drug and that the DFO on the fiber's surface dissolves rapidly into the water after the membrane is immersed in PBS. The release of the drug gradually slows down at subsequent times ( Figure 2E).

Evaluation of Biocompatibility of Tendon Anti-Adhesion Membrane
To explain the effect of the P(MMD-co-LA) membranes (R_ PDPLA, A_PDPLA, and A_PDPLA /DFO) on cell viability, the P(MMD-co-LA) membranes were co-cultured with tenocytes. Then the tenocytes are stained with Calcein-AM/PI (Figure 3A), and the results show no significant difference in the proportion of dead and alive cells between the P(MMD-co-LA) membranes group and the control group ( Figure 3C).

Oriented Fiber Membrane Regulates the Shape and Polarization of Macrophages
Macrophages were cultured on oriented and random P(MMD-co-LA) membranes to study the effects of material structure on the morphology and polarization of macrophages. The immunofluorescence staining shows that macrophages stretched randomly and showed disc shape on the R_PDPLA membrane, while on A_PDPLA and A_PDPLA/DFO membrane, they tend to extend in spindle shape along the orientation of nanofiber membrane ( Figure 3B). In addition, the expression levels of CD206, TNF-α, and IL-10 were quantitatively analyzed by RT-qPCR, in which the CD206 is a marker of M2 macrophages, and the TNF-α and IL-10 are mainly secreted by M1 and M2 macrophages. Our results show more CD206 expression in A_PDPLA and A_PDPLA/DFO membrane groups. The macrophages inoculated on A_PDPLA and A_PDPLA/DFO membranes secrete less TNF-α and more IL-10 than those on the R_PDPLA membrane and culture plate ( Figure 3D-F).

DFO (1%) Oriented Fiber Membrane Promotes HUVECs Migration and Tube Formation In Vitro
The formation of blood vessels in tendons involves two main processes: endothelial cell migration and angiogenesis. The endogenous healing of tendons with angiogenesis is the key to reducing tendon adhesion. To evaluate the effect of the A_PDPLA/DFO membrane on angiogenesis, the HUVECs scratch test was carried out. After 12 h of culture, nearly half of the scratches were closed in the A_PDPLA/DFO group, while those in the other groups were relatively low (Figure 4A,C). After 24 h of culture, the scratch gap of the A_PDPLA/DFO group was almost entirely closed, while that of the other groups was only nearly half of the healing area. The results showed that DFO could significantly promote the migration of HUVECs ( Figure 4A

www.advmatinterfaces.de
test was used to evaluate the effect of the A_PDPLA/DFO membrane on the angiogenesis of HUVECs. According to the experimental results, it is observed that compared with R_PDPLA and A_PDPLA membrane groups, A_PDPLA/DFO membrane group can significantly promote angiogenesis of HUVECs at both time points ( Figure 4B). After incubation for 3 h, several different tubules can be formed in the A_PDPLA/DFO membrane group, which are rarely seen in the R_PDPLA and A_PDPLA membrane groups. Subsequently, the number of tubules in the A_PDPLA/DFO membrane group gradually increases and becomes more apparent. Although the number of tubules in the R_PDPLA and A_PDPLA membrane groups increases gradually at 6 h, it is much less than that in A_PDPLA/DFO membrane group and equivalent to that in A_PDPLA/DFO membrane group at 3 h ( Figure 4D).

Effect of DFO on Angiogenesis of HUVECs
To explore the mechanism and pathway of DFO in promoting angiogenesis, we used an immunofluorescence test and quantitative RT-qPCR test to study the effect of DFO on HUVECs and to explore the expression of related factors of DFO in the process of promoting angiogenesis. The cellular immunofluorescence images show that compared with the R_PDPLA and A_PDPLA membrane groups, the expression of the HIF1-α gene in the A_PDPLA/DFO membrane group is significantly up-regulated, and the positive regions of HIF1-α, VEGF, and CD31 are also significantly increased in A_PDPLA/DFO membrane group (Figure 5A-C). In addition, in quantitative RT-qPCR analysis, the expression of HIF1-α, VEGF, and vascular maturity marker CD31 in the A_PDPLA/DFO membrane group is significantly higher than in other groups ( Figure 5D-F). However, there is no difference in the expression level of angiogenesis-related genes between the R_PDPLA and A_PDPLA membrane groups ( Figure 5D-F).

Improvement of Tendon Repair by Oriented Fiber Membrane Loaded with DFO
To evaluate the adhesion of tendons, gross observation was carried out 3 weeks after the operation. It was found that the  www.advmatinterfaces.de wounds healed well, and there was no infection or ulcer, and then the tendon repair site was exposed. It can be seen that there is dense adhesive tissue around tendon repair in the control group without fibrous membrane, and most of them need sharp weapon separation. In the experimental group wrapped with R_PDPLA membrane, more than half of the tendon and the surrounding adhesion tissue also need to be separated by sharp weapons. The A_PDPLA membrane group also shows that some fibrous tissue has adhered between the tendon and the surrounding tissue. At the same time, there is almost no adhesion around the oriented fiber membrane loaded with DFO ( Figure 6A). The results show significant differences in the degree of adhesion between the DFO-oriented fiber membrane group, the A_PDPLA membrane group, the R_PDPLA membrane group, and the control group ( Figure 6F).
To study the effect of DFO-oriented fiber membrane on tendon adhesion and healing at the tissue level, 3 weeks after the operation, the repaired tendon tissues were removed. Then the tissues were made into tissue sections and immunohistochemical staining. It is observed that there is severe adhesion and poor healing between the tendon and the surrounding granulation tissue in the control group. And most of the adhesion and poor healing are found in the R_PDPLA membrane group and partly in the A_PDPLA membrane group. However, there is almost no adhesion and good healing around the tendon in the DFO-oriented fiber membrane group ( Figure 6B). According to the results of statistical analysis, the representative tissue sections of tendons wrapped with A_PDPLA/DFO membrane are compared with those of the A_PDPLA membrane group, and the representative tissue sections of tendons wrapped with A_PDPLA membrane are compared with those of the control group. It is found that there are significant statistical differences in tendon adhesion and healing ( Figure 6G,H).
To evaluate the effect of the material on tendon healing and mechanical recovery, 3 weeks after the operation, the repaired tendon was removed. And the adhesive tissue on the surface of the tendon is removed and tested by the electronic universal tensile test machine ( Figure 6C). It is found that the maximum tensile strength of the oriented fiber membrane group loaded with DFO is significantly higher than that of the A_PDPLA membrane group, and the maximum tensile strength of the A_PDPLA membrane group is significantly higher than that of the control group and the R_PDPLA membrane group ( Figure 6I). According to the statistical analysis results, there is significant statistical significance in the maximum tensile strength between A_PDPLA/DFO membrane group and the

www.advmatinterfaces.de
A_PDPLA membrane group, the A_PDPLA membrane group and control group, and the R_PDPLA membrane group ( Figure 6I).
To further verify the effect of DFO-loaded oriented fibrous membrane on preventing tendon adhesion, a Western blot semi-quantitative test was used to evaluate the expression of collagen I and collagen III in the tissue around the repaired tendon after 3 weeks, and GAPDH was used as the protein internal control. We can directly see that the DFO-oriented fiber membrane group expresses less collagen I and collagen III than the A_PDPLA membrane group, while compared with the A_PDPLA membrane group, the control group has more expression of collagen I and collagen III, but the expression of the control group and R_PDPLA membrane group is almost the same ( Figure 6D). In addition, the results of RT-qPCR quantitative analysis confirm significant differences in the expression of collagen I and collagen III between the DFO-oriented fiber membrane group and A_PDPLA membrane group, A_PDPLA membrane group and control group. Still, there is no significant difference between the control group and the R_PDPLA membrane group ( Figure 6E).

DFO-Loaded Oriented Fiber Membrane Regulates the Polarization of Macrophages and Inhibits Oxidative Stress
To evaluate the ability of an oriented structure to regulate the polarization of macrophages, the CD68, CD206, CCR7, IL-10, and TNF-α were detected by immunofluorescence. In the process of tendon healing, the initial inflammatory stage is the beginning of tendon healing and adhesive tissue formation. After early acute inflammation, the macrophage group around the material will gradually change from classical activation (M1 phenotype) to alternative activation (M2 phenotype), which overlaps with the subsequent tendon proliferation stage. To evaluate the regulatory effect of fibrous membrane structure on inflammation and macrophage polarization in the treatment of promoting tendon healing and reducing tendon adhesion, the CCR7 and TNF-α were selected as markers and inflammatory markers of M1 macrophages, and the CD206 and IL-10 were selected as markers and antiinflammatory markers of M2 macrophages. As is shown in Figure 7A, by three-color immunofluorescence staining of macrophage polarization phenotypic genes, inflammatory genes, and anti-inflammatory genes expressed in tendon tissue 3 weeks after the operation, we can see that the proportion of CCR7/CD68 (M1 macrophage proportion) and TNFα expression in tendon tissue of A_PDPLA and A_PDPLA/ DFO membrane groups are lower than those of injured tendons repaired by R_PDPLA membrane. To further explore the effect of the immune microenvironment on the level of inflammation during tendon repair, quantitative RT-qPCR analysis was performed. The results show that the proportion of M1 macrophages and the level of TNF-α in tendon repair tissue in A_PDPLA and A_PDPLA/DFO membrane groups are significantly different from those in the R_PDPLA membrane group. There is also a significant difference in the level of TNF-α between the A_PDPLA and A_PDPLA/DFO membrane groups ( Figure 7C,D).
Since macrophages are the primary type of inflammatory cells in the early tendon healing process, their polarization to M1 or M2 phenotypes will have opposite biological effects. Therefore, we further studied the effect of orientation structure on the polarization of macrophages, and 3 weeks after the operation, the tendon sections were stained with CD68, CD206 and IL-10 immunofluorescence staining. As is shown in the representative fluorescence picture of Figure 7B, in the repaired tendon, the R_PDPLA membrane group shows a lower expression of CD206 and IL-10. In contrast, the A_PDPLA and A_PDPLA/DFO membrane groups show more CD206 and IL-10 favorable distribution, which indicates that the A_PDPLA and A_PDPLA/DFO membrane groups produce more M2 macrophages and a higher level of IL-10 expression at the repair site. Similarly, quantitative RT-qPCR analysis was performed to study further the mechanism of the effect of oriented fiber membrane on the repair of the injured tendon in regulating the regenerative immune microenvironment. The results show that the proportion of CD206/CD68 (M2 macrophage) and the expression level of IL-10 in tendon tissue of the A_PDPLA and A_PDPLA/DFO membrane groups are significantly different from those of the R_PDPLA membrane group ( Figure 7E,F). At the same time, to evaluate the free radical scavenging potential of the A_PDPLA/DFO membrane, we detected the expression of two antioxidant proteins, MnSOD and HO-1. RT-qPCR analysis shows that the expression of MnSOD and HO-1 in the A_PDPLA/DFO membrane group is significantly higher than that in A_PDPLA and R_PDPLA membrane groups ( Figure 7G).

Improvement of Vascular Remodeling Of Oriented Fibrous Membrane Loaded With DFO
Based on in vitro studies, immunofluorescence and RT-qPCR tests were used to study the effect of DFO released by A_PDPLA/DFO membrane on angiogenesis 3 weeks after wrapping tendons to explore further the role of A_PDPLA/DFO membrane in inducing rapid angiogenesis in vivo. We can see that the outline of the newly formed vascular network is visible in the immunofluorescence results of the A_PDPLA/DFO membrane group (Figure 8D). The results show that A_PDPLA/DFO membrane can significantly promote the formation of blood vessels in the repaired tendon. In addition, tissue immunofluorescence Adv. Mater. Interfaces 2023, 10, 2300054 images also show that the expression of HIF1-α in tissue is upregulated in the presence of DFO release ( Figure 8A). At the same time, the expression of HIF1-α targeting genes, SDF-1α and VEGF, and angiogenesis markers, CD31, are also up-regulated in A_PDPLA/DFO membrane group. And the positive areas of SDF-1α, VEGF, and CD31 in tissue immunofluorescence images increase significantly. It is suggested that the fibrous membrane loaded with DFO promotes the expression of genes related to angiogenesis in vivo ( Figure 8B-D). Next, we further discussed the expression of genes and proteins related to the role of DFO in promoting angiogenesis. After quantitative RT-qPCR analysis, the results show that the expression of HIF1-α, SDF-1α, VEGF and CD31 genes in the A_PDPLA/DFO membrane group is higher than that in the A_PDPLA membrane group and R_PDPLA membrane group ( Figure 8E,F). The combined study of immunofluorescence image results and RT-qPCR analysis results in 3 weeks after the operation showed that the A_PDPLA/DFO membrane promoted the expression of angiogenesis marker gene CD31 in tendon tissue. In contrast, CD31 expression was less in R_PDPLA and A_PDPLA membrane groups. These results show that the local release of DFO significantly enhances angiogenesis compared with other groups in vivo ( Figure 8A-F).

Discussion
In recent years, studies on tendon adhesion after an injury have emphasized that the functional recovery of the tendon is limited due to too much exogenous healing and too little endogenous healing. The main reason is the lack of effective regulation of the tendon-healing microenvironment. Among them, immune factors and angiogenesis are important factors affecting the formation of an excellent healing microenvironment in the tendon repair process. After a tendon injury, it is vital to avoid the abnormal migration of fibroblasts, so the fibrous membrane should be made into a barrier structure with a small pore diameter [15,37,43,46] to prevent fibroblasts around the tendon from invading the healing end of the tendon. [20,23,47] In addition, abnormal macrophage activity is the main driving factor of many kinds of tissue fibrosis during the inflammatory period. Among them, type I macrophages can secrete a variety of inflammatory cytokines and increase inflammatory response, while type II macrophages inhibit the inflammatory response and promote tissue repair. Therefore, it is necessary to regulate the polarization of macrophages. [35,48,49] Because tendon cells and tissues need nutritional support during endogenous healing, it is accompanied by the formation of blood vessels   www.advmatinterfaces.de migration, guide cell polarization, and angiogenesis, to inhibit tendon adhesion and provide nutritional support for endogenous healing, and finally achieve the effect of inhibiting tendon adhesion and promoting routine tendon healing.
Here, the P(MMD-co-LA) membranes were successfully synthesized by ring-opening polymerization, and oriented nanofiber membranes loaded with DFO were prepared by electrospinning. The fiber membrane had the characteristics of uniform pore structure and good mechanical properties, combined with the previous study of peripheral nerve regeneration [35] and the results of biosafety evaluation of tendon cells in vitro. The results show that the fiber membrane has good biocompatibility and can be used to prepare tendon anti-adhesion membranes. When the membrane is used to prevent tendon adhesion, it is found that the application effect of the A_PDPLA membrane is better than that of the R_PDPLA membrane, and the anti-adhesion effect of the A_PDPLA membrane is further improved after loading DFO. However, the specific mechanism of DFO-loaded oriented fiber membrane in preventing tendon adhesion is still unclear. Its long-term application effect remains to be further studied, However, we have shown that oriented fiber membrane combined with DFO can down-regulate the level of inflammatory factors, promote the expression of anti-inflammatory factors, increase the level of antioxidants and promote angiogenesis. These favorable healing microenvironments reduce tendon adhesion formation.
Our in vitro experimental results and previous studies showed that [35] the uniform longitudinal orientation of the membranes of A_PDPLA and A_PDPLA/DFO significantly affected the arrangement of macrophages. In addition, some studies have reported similar results that PCL with oriented or randomly oriented structures can change the shape and phenotype of cells during tendon tissue repair. [49] Moreover, some studies have proved a specific relationship between the secretion of inflammation-related cytokines and the polarization of macrophages and found that macrophages growing on oriented fibrous membranes can significantly reduce the secretion of pro-inflammatory factor TNF-α. Furthermore, other studies have demonstrated that macrophages can achieve an anti-inflammatory phenotype by elongating growth in fibers and grooves. [35,[49][50][51] At the same time, our study found that macrophages inoculated on the oriented fibrous membrane could significantly reduce the pro-inflammatory factor TNFα and promote the secretion of the anti-inflammatory factor IL-10. Our in vivo results showed that 3 weeks after the operation, the A_PDPLA membrane group increased the expression level of M2 macrophage marker CD206 and the corresponding anti-inflammatory level while decreasing the expression level of M1 macrophage marker CCR7 and the inflammatory level of the repaired tendon. In contrast, the R_PDPLA membrane group had a lower proportion of M2 macrophages and a higher proportion of M1 macrophages and secreted more inflammatory cytokines TNF-α and less anti-inflammatory cytokines, IL-10. Furthermore, oxidative stress is also vital for the healing microenvironment of damaged tendons. [7] When the fibrous membrane is wrapped in the damaged tendon, the foreign body reaction, local ischemia, and inflammation in the injured site will cause a considerable accumulation of ROS. Excessive ROS accumulation in the injured site and the relatively insuf-ficient endogenous antioxidants will lead to an oxidative tendon injury. [7,52,53] Therefore, it is necessary to regulate oxidative injury and inflammation to promote the functional recovery of tendons. The healing microenvironment of the fiber membrane is poor, so other controlled-release drugs should be considered in preparing the fiber membrane. [38,39] Since DFO can reduce the attack effect of ROS, its application in antioxidation is very important. [34,54,55] At the same time, our results showed that the significant expression of key antioxidant enzymes such as MnSOD and HO-1 was involved in regulating oxidative stress, which confirmed that the A_PDPLA/DFO membrane played a specific role in antioxidation. In addition, ROS can also lead to immune activation after congenital tendon injury, which further activates the inflammatory response. [56] Our results also showed that the expression of TNF-α, a marker of inflammation in the A_PDPLA/DFO membrane group, was significantly lower than that in the A_PDPLA membrane group. In conclusion, our results in vitro and in vivo showed that oriented fiber membrane significantly promoted the polarization of macrophages to M2 phenotype and down-regulated the secretion of pro-inflammatory factors. After adding DFO, inflammatory reaction and oxidative stress in the fibrous membrane decreased further.
After a tendon injury, tendon cells and tissues need nutritional support in endogenous healing, often accompanied by the formation of blood vessels in the tendon. Rapid vascularization plays a vital role in endogenous tendon healing because it can provide nutrients for tissue and cell growth and play an important role in the cellular and signal molecular regulation of tendon healing. [23,57,58] Under normoxic conditions, DFO is widely used as a simulated hypoxia compound. [59] It can promote the secretion of VEGF and SDF-1α in vascular injured tissue by blocking the degradation of HIF1-α, thus promoting cell proliferation and growth. [60] We studied the effect of DFO on promoting angiogenesis by targeting HIF1-α tissue healing to treat tendon injury and reduce tendon adhesion. Alexis et al. have reported that DFO accelerates vascularization by relying on the threshold, in which excessive DFO significantly reduces the time of angiogenesis. [36] Therefore, in our previous work, we evaluated the appropriate concentration of DFO loading in the fiber membrane, and the results indicated that the PDPLA/DFO fiber membrane containing a 1% concentration of DFO could produce enough DFO to meet the migration and tube formation of HUVECs in vitro. [35] In addition, when HUVECs were co-cultured with A_PDPLA/ DFO membrane, the secretion of HIF1-α increased, and the secretion of VEGF, which is closely related to cell migration and vascularization, also increased significantly. More importantly, to evaluate the activation of the HIF1-α pathway during tendon repair 3 weeks after A_PDPLA/DFO membrane implantation, the immunofluorescence and RT-qPCR tests were performed on the repaired tendons in each group. It was found that the fibrous membrane significantly promoted the secretion of SDF-1α and VEGF in the injured tendon, which was very important for promoting the migration of endothelial cells to the injured tendon and the formation of vascularization and further promoting the endogenous healing of the tendon. The results showed that the tendon tissue of the A_PDPLA/DFO membrane group was repaired well, and there www.advmatinterfaces.de were more blood vessels in the repaired tendon. These results clarified part of the molecular mechanisms of A_PDPLA/DFO promoting angiogenesis and tendon function recovery and indicated that local release of DFO could accelerate endothelial cell migration and vascularization and promote the construction of favorable healing microenvironment and tendon repair process.
To explore the effect of different fibrous membranes wrapping on reducing tendon adhesion and promoting tendon functional recovery, we evaluated the adhesion, healing, and functional recovery of tendons in each group by Macroscopic evaluation, histological evaluation, biomechanical evaluation and Western blot analysis of collagen I and collagen III in repaired tendon tissue. It was found that compared with Control and R_PDPLA membrane groups, the tendon tissue of the A_PDPLA membrane group had less adhesion, better healing, and stronger mechanical properties. Based on the A_PDPLA membrane group, the results showed that the combination of DFO and A_PDPLA membrane could more effectively reduce the formation of tendon adhesion, promote the routine healing of tendons, and improve the mechanical properties of tendons more effectively. To sum up, our results show that adjusting immune factors by oriented structure and loading DFO promotes vascularization to improve the tendon healing microenvironment, which has a specific effect on reducing tendon adhesion, promoting routine tendon healing and functional recovery.

Conclusion
This study constructed a new tendon anti-adhesion membrane with anti-inflammatory and vascularization properties by aligned P(MMD-co-LA) combined with DFO. This kind of fiber membrane had not been used to prevent tendon adhesion. The oriented P(MMD-co-LA) membrane could induce macrophages to polarize to the M2 phenotype and produce an anti-inflammatory effect. In addition, after the DFO was loaded into the fibrous membrane, the in situ release of DFO could effectively induce angiogenesis and provide nutritional support for the endogenous healing of tendons. In summary, these findings suggested that the oriented fiber membrane loaded with DFO could improve the microenvironment of tendon healing by promoting vascularization and regulating the anti-inflammatory properties of macrophages and had a specific effect on reducing tendon adhesion.