Antibacterial and angiogenic dual-functional fibrous membrane dressing for infected wound healing

Background: Skin, being a vital organ that regulates physiological responses in the human body, is prone to injury from external environmental factors. Healing full-thickness skin defects becomes especially challenging when infections and vascular injuries are involved. Traditional wound dressings with single functions, such as antibacterial or angiogenic properties, fall short in achieving rapid wound healing. To address this, there is a need to develop wound dressing materials that possess both effective antibacterial and angiogenic properties. Methods: In this study, we utilized electrospinning technology to fabricate hyaluronic acid -cellulose acetate fibrous membrane dressings, incorporating poly(ionic liquid) as an antibacterial polymer and deferoxamine as an angiogenic agent. Results: The resulting fibrous membrane dressing contained poly(ionic liquid) and deferoxamin showcased a microporous structure, drug-releasing capabilities, and excellent air permeability. It not only demonstrated highly effective antibacterial properties but also exhibited remarkable angiogenesis, thereby promoting the healing of full-thickness skin defect wounds in both in vitro and in vivo assays. Conclusion: These findings highlight the immense potential of this wound dressing material for future clinical applications.


Background
The treatment of full-thickness wounds poses a significant burden on public resources and expenses, especially with the rising incidence of infected wounds in an aging global population.Full-thickness defect wound is difficult to heal because bacterial infection and subsequent inflammatory reactions, as well as vascular disorders at the wound site leads to a poor immune response [1][2][3][4].Wound dressing materials can provide a barrier to the outside world, which not only avoids infection but also provides a conducive environment for skin tissue repair and regeneration [5][6][7].In order to expedite the wound healing process for infected wounds, research have been focused on studying wound dressings with excellent antibacterial, anti-inflammatory and angiogenic properties [8][9][10].Various types of wound dressing materials, including hydrogels, membranes, and sponges, have been studied, with fibrous membranes, in particular, offering high porosity and surface area for effective exudate absorption from the wound [11][12][13][14][15][16][17][18].
Various antibacterial fibrous membranes dressings have been developed via electrospinning technology.Wang et al. used polycaprolactone and polyvinylpyrrolidone as matrix materials, and magnesium oxide nanoparticles as antibacterial agents to generate a MgO/PCL/PVP antibacterial nanofiber membrane [19].Cheng et al. developed polyacrylonitrile antibacterial electrospun fibrous membrane, using silver nanoparticles as an antibacterial agent.The membrane was able to release silver nanoparticles constantly and was found to exhibit long-lasting antibacterial property [12].Salih et al. fabricated polylactic acid electrospun fibrous membrane loaded with various antibiotics such as vancomycin, ceftriaxone, and cefoperazone.The membrane loaded ceftriaxone was found to demonstrate activity against S. aureus and E. coli, while the membrane loaded cefoperazone and vancomycin was found to inhibit the growth of S. aureus [20].In the recent years, poly(ionic liquids) as a type of high-performance antibacterial polymers have also been widely studied.Its increasing popularity can be attributed to their biocompatibility and low multi-drug resistance [21].Mao et al. recently designed an ionic liquid 3-hexyl-1-vinylimidazolium bromide copolymerized with 2,2'-(5-acrylamido-1-carboxypentylazanediyl) diacetic acid and acrylonitrile [22].The synthesized copolymer was complexed with cerium (IV) ions to form an antibacterial membrane by electrical spinning.The obtained poly (ionic liquid) (PIL)-Ce membrane dressing demonstrated excellent antimicrobial properties and cleavage activities to antibiotic-resistant gene in in vitro and in vivo assays.
Despite these advancements, traditional single-functional electrospun fibrous membranes still fall short in meeting the requirements for healing full-thickness skin defect wounds [28].Thus, there is a demand for dual-functional electrospun dressings with both exceptional antibacterial and angiogenic properties.In this study, we have developed a hyaluronic acid (HA)-cellulose acetate (CA) electrospun fibrous membrane dressing loaded with halogen-free imidazolium poly(ionic liquids) and DFO.Through in vitro and in vivo investigations, we explored its antibacterial activity, angiogenic potential, and wound healing properties (Figure 1).This novel dual-functional dressing holds promise for advanced wound care applications.

Synthesis of halogen-free imidazolium poly(ionic liquids)
The synthesis of poly(butyl imidazole acetate)(PBIA) involved several steps.First, 1,4-butanediamine (0.0225 mol) was thoroughly mixed with 11.25 mL of deionized water.Then, acetic acid (0.045 mol) was added dropwise and stirred evenly.Next, under ice bath conditions, ethylene glycol (0.0225 mol) and formaldehyde (0.0225 mol) were added to the solution.After adding all the reagents, the ice was removed, and the pH of the solution was adjusted to 7. The reaction temperature was then raised to 100 °C.During the reaction, the color of the solution changed from transparent to golden, and eventually to black.The reaction was allowed to proceed for approximately 36 h.Once the reaction was completed, the sample underwent purification by dialysis for 18 h.After dialysis, the sample was frozen at −80 °C and then dried.The other four imidazolium poly(ionic liquids)poly(butyl imidazole propionate) (PBIP), poly(butyl imidazole butyrate) (PBIB), poly(butyl imidazole pentanoate) (PBIV), and poly(butyl imidazole hexanoate) (PBIC) were synthesized using a similar method.Refer to Supplementary Figure S1 S2E).

Preparation of electrospun fibrous membrane by electrospinning
To prepare the fibrous membrane, hyaluronic acid (0.15 g, 1 × 10 −5 mM), PBIC (0.015 g, 1×10 −2 mM), CA (3.6 g, 2.4×10 −4 mM), and DFO (0.015 g, 2.68×10 −2 mM) were completely dissolved in a mixture of acetone (21 mL), dimethylacetamide (DMAC) (9 mL), and water (3 mL) to form a spinning solution.A tin foil was placed on the receiver to facilitate the collection of the spinning fibrous material.The spinning needle (diameter = 0.81 mm) was fixed on a syringe, and the distance between the needle and the receiver was set to 13 cm.The ambient temperature and humidity were maintained at 30 °C and 50%, respectively.The spinning solution (10 mL) was drawn at a pushing speed of 0.01 mm/s, and the voltage was set to 27 kV.Subsequently, the electrospun fibrous membrane was dried at 37 °C and named as HA-CA-PBIC-DFO.The fibrous membrane without DFO was named as HA-CA-PBIC, and the one without both PBIC and DFO was named as HA-CA for comparison.

Structure and chemical & physical properties characterization of halogen-free imidazolium poly(ionic liquids) and electrospun fibrous membranes
Characterization of halogen-free imidazolium poly (ionic liquids).The structure of halogen-free imidazolium poly(ionic liquids) was characterized by 1H NMR (AVANCE III 400MHz, Germany) , GPC (Waters 1515, USA).Zeta potential and particle size were characterized by laser particle size analyzer (ZEN3600, UK).Surface morphology characterization of electrospun fibrous membranes.The surface morphology of electrospun fibrous membranes were observed by FE-SEM (Zeiss Sigma 500, Germany) and AFM (Dimension, Germany).The permeability of electrospun fibrous membranes.Water (10 mL) and electrospun fibrous membranes (3 × 3 cm) were placed in vial bottles (diameter = 1.5 cm).The bottles were covered with the fibrous membranes and then sealed with parafilm.For comparison, vial bottles without any covers were used as the control.All the vial bottles were weighed before being placed in an oven, where they were heated to 37 °C for a duration of 24 h.After the heating process, the glass bottles were removed from the oven and weighed again.The permeability of the electrospun fibrous membrane was calculated using Equation (1): Where m 1 is the initial weight of vial bottles, and m 2 is the weight of vial bottles after 24 hours.

In vitro release testing of electrospun fibrous membranes
The release rate of the sample over time can be tested using absorbance.The antibacterial electrospun fibrous membrane containing acetic acid was placed in water, and the release rate of poly(ionic liquids) within 24 h was measured using the UV-visible spectrophotometer (UV-1500, Shimadzu, Japan).

Minimum inhibitory concentration (MIC) testing of poly (ionic liquids)
MIC is used to assess the inhibitory effect of the samples on bacteria.Firstly, poly(ionic liquids) were prepared in PBS at a concentration of 500 µg/mL.Next, E. coli and S. aureus were cultured to a concentration of 106 CFU/mL.A 96-well plate was then filled with 100 µL of LB and MH liquid culture media.Subsequently, 100 µL of the bacterial suspension and 100 µL of the poly(ionic liquids) were added to the first well and serially diluted across the plate.The plate was placed in a biochemical incubator and incubated for 24 hours.After incubation, the OD values of each well were measured at 600 nm using an enzyme-labeled instrument, as shown in Equation ( 2): Where d (OD) is the optical density value of the PBS blank control group, and d' (OD) is the optical density value of the five polyelectrolyte liquid samples.

In vitro of cell viability of poly (ionic liquids) and electrospun fibrous membranes
Cell-toxicity is a crucial parameter for evaluating the clinical applicability of antimicrobial dressings.The CCK-8 method was employed to measure the cell toxicity of the samples.Initially, poly(ionic liquids) were prepared and dissolved in PBS to a final concentration of 12 mg/mL, followed by ultraviolet sterilization.Next, the sample solution was mixed with an equal volume of HSF at a concentration of 5 × 10 4 cells/mL.The mixture was then placed in a cell culture incubator for 24 hours.After the incubation period, 20 µL of CCK-8 detection solution was added to the mixture, and further incubated for 2 h.Subsequently, the absorbance of the solution was measured at 450 nm.A PBS solution was used as the blank control group.
For the electrospun fibrous membranes were sterilized by UV irradiation for 30 min.Following this, they were soaked in a PBS solution for 24 h, and the extraction solution was filtered through a 0.22 μm filter membrane.The experimental procedure for the extraction solution was similar to the previous method.The cell survival rate was calculated using Equation (3): Where d (OD) is the optical density value of PBS of the blank control group; d' (OD) is the optical density value of the samples.

Hemolytic assay of electrospun fibrous membranes
To assess the hemolytic activity of hydrogels, fresh human blood (5 Submit a manuscript: https://www.tmrjournals.com/bmecmL) was obtained from a healthy 21-year-old male.The blood was then centrifuged at 1,200 rpm for 8-10 min using a high-speed centrifuge to separate the red blood cells from the blood plasma.The supernatant was carefully aspirated, and the red blood cells were washed with PBS buffer three times, with each wash involving a new centrifugation step.Following the washing process, a 5% v/v red blood cell solution was prepared using PBS.Once the blood was processed, the samples were prepared for testing.The extraction of electrospun fibrous membranes (100 µL, extracted for 24 h) were placed in each well of a 96-well plate, and 100 µL of the red blood cell solution was added to each well.To ensure adequate contact between the sample and the cell solution, the plate was placed on a constant-temperature shaker and shaken at 200 rpm for 1 h.After incubation, the plate was centrifuged again for 8-10 minutes, and the supernatant was collected.Finally, the absorbance at 540 nm was measured using a microplate reader (Infinite F50).A red blood cell solution dissolved in 0.3% Triton X-100 was used as a positive control, and a red blood cell solution without any sample served as a negative control.Each experimental group was measured seven times.The hemolysis percentage was calculated using Equation (4): Where OD sample is the absorbance of the samples, OD positive is the absorbance of the positive control group, and OD negative is the absorbance of the negative control group.

In vitro antibacterial performance testing
The electrospun fibrous membranes were subjected to sterilization by UV irradiation for 30 min.Afterward, they were cut into thin slices of the same size (1 cm × 1 cm) and placed in a 24-well plate.To each well, 1 mL of E. coli (6 × 10 8 CFU) or S. aureus (9 × 10 8 CFU) suspension was added.The plate was then placed in a biochemical incubator for incubation with the samples and bacterial suspension for 2, 24, and 72 h, respectively.
Upon reaching the designated incubation time, the samples were removed from the wells, and the fibrous membrane was washed thrice with sterile water.Subsequently, 10 mL of sterile water was added to a sterilized centrifuge tube.The fibrous membrane was placed into the centrifuge tube, and an ultrasonic cleaner was used for 30-40 min to disperse the bacteria adsorbed onto the surface of the sample into the sterile water.From the final diluted solution, a 100 mL aliquot was taken and spread onto an LB or MH solid culture medium.The plate was then inverted and incubated for 24 h in an incubator.The log-reduction value of the bacteria was calculated using Equation ( 5): CFU control is the colony of HA-CA, and CFU sample is the colony of HA-CA-PBIC and HA-CA-PBIC-DFO.

In vitro anti biofilm activity assay
Separate glass dishes (l = 24 mm × 24 mm) were inoculated with E. coli (6 × 10 8 CFU) and S. aureus (9 × 10 8 CFU) suspensions, respectively, and incubated for 24 h to allow biofilm formation.Once the biofilms had formed, sterilized membranes (1 cm × 1 cm) were carefully placed on top of each biofilm and further incubated for an additional 24 h.
After this incubation period, the membranes were removed, and the biofilms were subjected to staining using a live/dead staining kit.The staining process involved the use of 6 μM SYTO 9 and 30 μM propidium iodide, and it was carried out in a dark environment for 15 minutes.Following staining, the live/dead status of the bacteria within the biofilms was observed using a fluorescence confocal microscope (Zeiss LSM 710, Germany) at an excitation wavelength of 490 nm and an emission wavelength of 635 nm.
In vitro anti-inflammatory assay U937 cells were cultured and added to a 6-well plate.Subsequently, negative control, positive control, and sample groups were prepared.The positive control and sample groups were treated with LPS (2 μg/mL) and incubated in a cell culture incubator for 3 h.After the initial incubation, 5 mM ATP (10 μg/mL) was added to both the positive control and sample groups and co-incubated for 30 min.Next, sterilized membrane extracts (100 μL) were introduced to the sample group, and co-incubation continued for an additional 24 h.
Following the incubation period, RNA extraction was carried out as follows: First, the cell suspension from the three groups was centrifuged.Then, the culture medium was removed, and 1 mL of a total solution was added and mixed with the cells.The cell solution was combined with 200 μL of chloroform and subjected to centrifugation.The resulting supernatant was mixed with 500 μL of isopropanol and centrifuged for 8 min.Subsequently, the supernatant was discarded, and 1 mL of 70% ethanol was added to the solution, followed by another 3 min centrifugation.The supernatant was collected, and the centrifuge tube was placed in an elevated position to allow residual ethanol to evaporate gradually.After the ethanol evaporation was complete, 10 μL of DEPC water was added to the centrifuge tube, and RNA was dissolved by gentle tapping.
The concentration of RNA was then measured, and a special reagent kit was used for reverse transcription of messenger RNA and cDNA synthesis.Subsequently, the target gene was amplified by qPCR.The primer sequences utilized in the process can be found in Supplementary Table S1.

Fluorescence staining imaging assay
In a culture dish, 1 mL of culture medium was added, and human skin fibroblasts (100 μL, 5 × 10⁴ cells/mL) were mixed with 50 μL of the sample in the culture medium.The mixture was then incubated for 24 h.Subsequently, the culture dish was removed from the incubator, and the culture medium was aspirated using a pipette.The dish was washed once with PBS.Next, 2 mL of 4% paraformaldehyde was added, followed by its removal, and the dish was washed three times with PBS.To permeabilize the cells, 1-2 mL of 0.5% Triton X 100 was added to the culture dish and left for 15 min.The dish was then washed three times with PBS.Afterward, 1% BSA was added to the dish and incubated for 30 min.The BSA solution was removed, and the dish was washed with PBS to eliminate any remaining BSA solution.For cell staining, the cells were exposed to rhodamine-labeled ghost spider peptide for 1 h, and an equal volume of 4,6-diamidino-2-phenylindole reagent was added and allowed to incubate for 3 min.The final staining step was performed under low-light conditions to avoid photobleaching.Following staining, the cells were photographed to observe their growth and survival status.The procedure for HUVEC was similar to that of HSF.

Transwell assays
To further verify the effect of DFO on cell migration, transwell assays were performed on electrospun fibrous membranes using HSF and HUVEC cells.As the electrospun fibrous membrane cannot be dissolved in water, the extraction solution was used for the experiment.The experimental procedure was similar to the previous experiments.The cells' vertical migration was photographed using an inverted microscope.

Vascular formation assay
Matrigel (300 μL) was slowly added into the well plate and incubated for 30 min until the gel solidified.Subsequently, cultured HUVEC cell suspension (500 μL) was added.500 μL of the extract of the sterilized membranes (1 cm × 1 cm) were then added, and three parallel samples were made for each type of spun fibrous.After the addition of samples, the well plate was placed in the incubator overnight.Pictures of the tubes were taken using a live cell imaging.The length of the tube formed by the cells, and the number of cell rings were measured using Image J.

Gene expression and protein expression testing of related angiogenic factors
HUVEC cells were cultured in a 6-well plate.For the positive control and sample groups, LPS (2 μg/mL) was added in the same volume as Submit a manuscript: https://www.tmrjournals.com/bmecthe cell medium and incubated for 3 h.Following that, 5 mM ATP (10 μg/mL) was added and incubated for another 30 min.Afterward, 100 μL of extracted sterilized membranes (1 cm × 1 cm) was added and incubated for an additional 24 h.
RNA extraction was performed as follows: cell liquids were centrifuged, and the cell medium was discarded.Next, 1 mL of the total solution was added to the cell mixture and mixed with chloroform (200 μL) in another centrifuge tube.After centrifugation, the supernatant was mixed with isopropanol (500 μL) and centrifuged again for 8 min.The supernatant was then removed, and 1 mL of ethanol (70%) was added.The mixture was centrifuged for 3 min, and the supernatant was extracted.The centrifuge tube was placed in an elevated position, allowing the residual ethanol to evaporate slowly.Then, DEPC (10 μL) was added to the centrifuge tube, and RNA was dissolved by gentle tapping.The RNA concentration was measured, and a special reagent kit was used for reverse transcription of messenger RNA, cDNA synthesis, and target gene amplification through qPCR.The gene sequences are provided in Supplementary Table S1.
To measure the protein expression levels of VEGF and HIF-1α, Western blotting was employed.HUVEC proteins were extracted by electrophoresis and transferred onto three types of silk fibrous membranes.The membranes were then blocked for 1 h and incubated with primary antibodies at 4 °C for 12 h.Subsequently, the membranes were washed five times with PBST and incubated with secondary antibodies for 1 h.After washing the membranes, testing was performed.

In vivo assay of electrospun fibrous membranes for wound healing
The experimental procedure received approval from the institutional animal care and use committee of Changzhou University (Approval number: 20230310042).8-week-old Sprague Dawley® male mice (250 g) were purchased from Chang Zhou Cavens Laboratory Animal Co., Ltd, Jiangsu, China.
For the in vivo assessment of antibacterial activity and wound healing of hydrogels, two full-thickness skin wounds (diameter = 10 mm) were created on each side of the mice spine using biopsy punches after shaved and sterilized and sterilized using 75% ethanol, and then five replicates of electrospun fibrous membranes (HA-CA, HA-CA-PBIC, and HA-CA-PBIC-DFO) were applied onto the wounds and secured with elastic adhesive bandages.A wound without any membrane served as the control group.To avoid to damage the wound, the membranes were cut into 15 mm × 15 mm to cover the wound completely.The experiment was conducted over a period of 14 days, during which the membranes were changed, and the wound areas were measured at fixed intervals.On the 14th day, the mice were euthanized, and tissue samples from the wounds and surrounding areas were collected and cut into several pieces.These samples were then fixed in 4% paraformaldehyde overnight, embedded in paraffin, and sectioned (5 μm thick).The length of the new epidermis was assessed to evaluate the wound healing progress using immunofluorescence and immunohistochemical assays.

Statistical analyses
For comparison, experimental data were processed using one-way analysis of variance followed by Tukey's multiple comparison tests.Error bars were exhibited as mean ± SD. * P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.001 were considered statistically significant.

Preparation of halogen-free imidazolium poly(ionic liquids) as antibacterial polymer
Five halogen-free imidazolium poly(ionic liquids) were successfully synthesized using the poly-Radziszewski reaction [29].These poly(ionic liquids) included PBIA, PBIP, PBIB, PBIV, and PBIC, with molecular weights ranging from 1,000 to 1,500 (Table 1).The 1H NMR spectra of these halogen-free imidazolium poly(ionic liquids) (Supplementary Figure S2) revealed characteristic chemical shifts at 4.27 ppm and 7.42 ppm, corresponding to the ethylene protons connected to the quaternary nitrogen atoms and the imidazolium ring, respectively [30].Additionally, chemical shifts at 0.9 ppm and 1.0-2.1 ppm corresponded to the methylene protons and ethylene protons of the carboxylic acid, respectively.These spectral results confirmed the successful synthesis of the five different imidazolium poly(ionic liquids) with distinct structures.
Furthermore, the particle size of the five imidazolium poly(ionic liquids) (Supplementary Figure S3A) ranged from 600 nm to 950 nm, with increasing hydrophobic ethylene segments from carboxylic ions.The zeta potential of the imidazolium poly(ionic liquids) also increased from 2.3 mV to 12.5 mV (Supplementary Figure S3B) due to the increasing ethylene segments contributed by carboxylate.Notably, a higher zeta potential of the imidazolium poly(ionic liquids) correlated with a lower MIC against E. coli and S. aureus (Table 1).Furthermore, the biocompatibility of the five imidazolium poly(ionic liquids) was assessed through the CCK-8 assay, and all of them exhibited cell viabilities above 90%, indicating their biocompatibility (Table 1).Considering both the antibacterial activity and biocompatibility, PBIC was selected as the preferred antibacterial polymer to be incorporated into the subsequent electrospun fibrous membrane dressings.

Surface morphology and roughness of electrospun fibrous membrane dressings
The antibacterial and angiogenesis fibrous membrane dressing (HA-CA-PBIC-DFO) was prepared using a mixed solution of HA, PBIC, CA, and DFO, with a spinning speed of 0.01 mm/s and 27 kV.For comparison, the electrospun fibrous membrane dressings without DFO were designated as HA-CA-PBIC, while the membrane without both PBIC and DFO was named HA-CA.Figure 2A displays the surface morphology and photographs of HA-CA, HA-CA-PBIC, and HA-CA-PBIC-DFO, respectively.The HA-CA membrane exhibited a smooth nano fibrous structure, with an average diameter of 75 nm (Figurte 2B).Additionally, the average pore size of the HA-CA membrane was measured to be 1440 nm, and its air permeability was found to be 90.8%(Supplementary Figure S4A).
Upon incorporating the PBIC antibacterial polymer, the average diameter of fibers in HA-CA-PBIC decreased to 21 nm.This reduction was attributed to the changes in concentration and viscosity of the HA/CA/PBIC mixed polymer solution after introducing the positively charged PBIC particles [31].The porous size of the HA-CA-PBIC membrane also decreased, reaching 270 nm, and a high distribution of nano fibers was observed.As a result, the air permeability of HA-CA-PBIC decreased to 87.6%.With the addition of DFO to the mixed polymer solution, the average fiber diameter thickened to 33 nm in HA-CA-PBIC-DFO.The air permeability of HA-CA-PBIC-DFO was measured at 87.4%, and a similar fibrous distribution was observed in the membrane.Figure 3 illustrates the roughness of the   fiber membrane surface as measured by AFM.The HA-CA membrane appeared relatively smooth with an average roughness (Ra) of 23 nm.However, both HA-CA-PBIC and HA-CA-PBIC-DFO membranes exhibited higher Ra values, measuring 49 nm and 56 nm, respectively.This increase in roughness was attributed to the presence of PBIC particles complexed within the membrane.

In vitro antibacterial activity, anti-biofilm and anti-inflammatory assays of fibrous membrane dressings
The antibacterial activity of electrospun fibrous membranes plays a crucial role in the healing of infected wounds [32].The membrane dressings containing PBIC antibacterial polymer exhibited high antibacterial efficacy against E. coli, as shown in Figure 4A, with the killing ratio of HA-CA-PBIC reaching 99.67%, and HA-CA-PBIC-DFO reaching 99.18% compared to HA-CA within 2 h.Similarly, the antibacterial membrane dressings also demonstrated excellent activity against S. aureus, as depicted in Figure 4B, with the killing ratio of HA-CA-PBIC reaching 99.32% and HA-CA-PBIC-DFO reaching 99.26% within 2 h.Moreover, the antibacterial activity persisted for 27 h, with the killing ratios of the membranes (HA-CA-PBIC and HA-CA-PBIC-DFO) against E. coli and S. aureus remaining above 97%.
The highly effective and sustained antibacterial activity of the fibrous membrane dressings was attributed to the low MIC of PBIC and its release rate from the membranes over 24 h (Supplementary Figure S4B).
To gain further insights into the antibacterial mechanism of the fibrous membranes, E. coli and S. aureus biofilms (green fluorescence signal) were observed after treatment with HA-CA.However, the green biofilm signal disappeared, and was replaced by a red signal when treated with HA-CA-PBIC and HA-CA-PBIC-DFO.This indicated that most of the bacteria in the biofilm were killed, and the biofilm was dissolved (Figure 5).These results revealed that the antibacterial fibrous membranes not only effectively released PBIC to eliminate bacteria but also possessed antibiofilm properties through contact killing, thereby inhibiting biofilm formation.
Inflammation is a common occurrence in infected wounds.As shown in Figure 4C, the gene expression of pro-inflammatory cytokines (TNF-α, IL-6, and IL-1β) in human tissue lymphoma cells Submit a manuscript: https://www.tmrjournals.com/bmectreated with the extraction of HA-CA-PBIC and HA-CA-PBIC-DFO for 24 h was significantly reduced compared to those treated with HA-CA alone.Moreover, HA-CA-PBIC-DFO exhibited even better anti-inflammatory properties than HA-CA-PBIC, as both the released PBIC from the membrane and DFO from HA-CA-PBIC-DFO within 24 h contributed to reducing inflammation (Supplementary Figure S4B) [33].

In vitro biocompatibility of fibrous membrane dressings
The cell cytocompatibility of the electrospun fibrous membrane dressings was assessed using HSF cells and HUVECs through the CCK-8 assay.After incubation with the extracted electrospun fibrous membrane dressings for 24 h, the cell viability of HSF cells and HUVECs was calculated.The results demonstrated that the electrospun fibrous membrane dressings were non-toxic, as the cell viability of HA-CA, HA-CA-PBIC, and HA-CA-PBIC-DFO were all above 90% for both HSF and HUVECs (Supplementary Figure S6A).The release of PBIC from HA-CA-PBIC and HA-CA-PBIC-DFO did not negatively affect the cell viability of HSF cells and HUVECs.Furthermore, Supplementary Figure S5B showed that the hemolysis of cells induced by the fibrous membrane dressings was below 0.8%.According to the American Society for Testing and Materials (ASTM) F756-00 standard, a hemolysis degree of 2% or less is considered nonhemolytic for biomaterials [34].Therefore, the fibrous membrane dressings demonstrated excellent hemocompatibility.
In addition to cytocompatibility, the migration of HSF and HUVEC cells with the extracted fibrous membrane dressings was analysed using a transwell assay (Figure 6).Compared with the other two electrospun fibrous membrane dressings (HA-CA and HA-CA-PBIC), more HSF and HUVEC cells were observed to penetrate the transwell membrane in the presence of the extracted HA-CA-PBIC-DFO.This can be attributed to the lower toxicity of PBIC and DFO, as well as their ability to promote cell migration, creating a more favorable environment for cell movement and proliferation.

In vitro angiogenesis assay of fibrous membrane dressings
Angiogenesis plays a crucial role in promoting wound healing [35].To evaluate the angiogenic potential of the prepared hydrogel using Matrigel, we conducted a tube formation assay.After 12 hours of incubation with the extracted electrospun fibrous membranes, vessel networks were visibly formed (Figure 7A).Among the samples, HUVECs cultured with the extraction of HA-CA-PBIC-DFO showed the highest number of tube meshes (83) and longer tubes (8.5 μm) compared to HA-CA and HA-CA-PBIC (Figure 7B, 7C).This  enhancement was attributed to the release of DFO from the HA-CA-PBIC-DFO membrane, which promoted HUVECs migration and the formation of new blood vessels [36,37].
To further explore the in vitro vascular repair ability of the fibrous membrane dressings, we evaluated the relative gene and protein expressions in HUVECs.Notably, certain cytokines such as HIF-1α and Submit a manuscript: https://www.tmrjournals.com/bmecVEGF are closely associated with angiogenesis in human dermal fibroblasts [38,39].After 24 hours of incubation with the extracted fibrous membrane dressings, HA-CA-PBIC-DFO showed higher HIF-1α and VEGF gene expressions compared to HA-CA and HA-CA-PBIC (Figure 7D).Particularly, HIF-1α was more significantly up-regulated than VEGF in cells treated with HA-CA-PBIC-DFO, as HIF-1α acts as a downstream signal in the nucleus [40,41].This up-regulation of HIF-1α and VEGF in cells exposed to DFO released from HA-CA-PBIC-DFO was further confirmed by images of western blot (Figure 7E).

In vivo full-thickness skin defect wound healing by fibrous membrane dressings
To evaluate the actual effect of promoting wound healing, we utilized a mouse model with S. aureus-infected full-thickness skin defects.In this study, HA-CA, HA-CA-PBIC, and HA-CA-PBIC-DFO were chosen as experimental groups, while the wound without any treatment as the control group (Figure 8A).To simulate the continuous release of PBIC and DFO from the fibrous membrane dressings, we replaced the dressings every 2 days.The results showed a substantial reduction in the number of S. aureus colony-forming units (CFU) in wounds treated with HA-CA-PBIC and HA-CA-PBIC-DFO compared to the control (Figure 8B).The killing ratio of HA-CA-PBIC and HA-CA-PBIC-DFO against S. aureus after 14 days was 94.6% and 95.1%, respectively, which aligned with the in vitro antibacterial activity.Moreover, the wound healing ratio of the experimental groups (HA-CA, HA-CA-PBIC, and HA-CA-PBIC-DFO) after 14 days was notably higher than the control group (3.2%), with percentages of 40.1%, 77.4%, and 97.1%, respectively (Figure 8C).
Immunohistochemical (IHC) stained images of IL-6, HIF-1α, and VEGF at 14 days were shown in Figure 9.The HA-CA-PBIC-DFO group exhibited the lightest staining colour of IL-6, indicating effective anti-inflammatory properties due to the release of PBIC and DFO.Moreover, the staining colour of HIF-1α and VEGF was the most intense in the HA-CA-PBIC-DFO group, suggesting high expression levels of HIF-1α in HUVECs, which in turn induced more angiogenesis and enhanced blood vessel formation.These in vivo results collectively indicated that the DFO released from the HA-CA-PBIC-DFO membrane exhibited a significant pro-angiogenesis effect.

Discussion
As we known the wound healing procession had four successive stages: hemostasis, inflammation, proliferation, and remodeling.However, most of paper reported fibrous membrane dressings just only showed antibacterial property which were limited to apply in full-thickness wound healing.Taken together, we created dual-functional fibrous membrane (HA-CA-PBIC-DFO) dressing by incorporating DFO and PIL into the HA-CA fibrous membrane, which showed antibacterial and angiogenesis to repair full-thickness skin defect wound.In the early stage (2 h), the dual-functional fibrous membrane can antibiofilm by contact-killing due to the high effective antibacterial PIL released (Supplementary Figure S4B), then DFO was released from HA-CA-PBIC-DFO membrane to exhibit anti-inflammatory and promote angiogenesis at the second and third stage of wound healing.Although dual-functional fibrous membrane (HA-CA-PBIC-DFO) was developed for full-thickness wound healing, the defect of the HA-CA-PBIC-DFO membrane was also optimized in next research.Especially, the releasing ratio of DFO from the membrane was very close to that of PIL, which could reduce the anti-inflammatory and angiogenesis effect.Overall, we created a dual-functional fibrous membrane (HA-CA-PBIC-DFO) dressing repairing full-thickness skin defect wound, providing a new strategy for rapid wound healing in clinical application.

Conclusion
In summary, we designed various of halogen-free imidazolium poly(ionic liquids) (PBIA, PBIP, PBIB, PBIV and PBIC) and they exhibited excellent antibacterial property and biocompatibility.Subsequently, an electrospun fibrous membrane dressing (HA-CA-PBIC-DFO) was fabricated using a HA/CA mixed polymer, loaded with PBIC as an antibacterial polymer and DFO as an angiogenic pharmacy.The fibrous membrane dressings exhibited uniform morphology, air permeability and drug releasing properties.The electrospun fibrous membrane dressing exhibited highly effective long-term antibacterial activity (killing ratio against E. coli was 99.18% and against S. aureus was 99.26% against in 2 h, killing ratio of bacteria was above 97% in 27 h), biocompatibility (promoting cell migration), anti-inflammatory (down-regulating TNF-α, IL-6 and IL-1β) and angiogenesis (blood vessel tube formation) properties in in vitro assays.In addition, the antibacterial and angiogenic fibrous membrane also promoted full-thickness skin defect wound healing in vivo assay, which could be applied clinically in future.

Figure 1
Figure 1 Antibacterial activity, angiogenic potential, and wound healing properties.(A) Schematic of electrospun fibrous membrane dressings fabrication.(B) Its application in full-thickness skin defect wound healing.

Figure 8
Figure 8 In vivo full-thickness skin defect wound healing by fibrous membrane dressings.(A) Photographs of the wound healing process treated with fibrous membrane dressings.(B) In vivo assay of viable skin surface colonies of S. aureus after 14 days.(C) The wound healing ratio after 14 days treatment (n = 5, * P ≤ 0.05).HA, hyaluronic acid; CA, cellulose acetate; DFO, deferoxamine; PBIC, poly(butyl imidazole hexanoate).