In situ electrospun aloe-nanofiber membrane for chronic wound healing

Abstract

Alleviating excessive inflammation while accelerating chronic wound healing to prevent wound infection has remained challenging, especially during the coronavirus disease 2019 (COVID-19) pandemic caused by SARS-CoV-2 when patients experienced difficulties with receive appropriate healthcare. We addressed this issue by developing handheld electrospun aloe-nanofiber membranes (ANFMs) with convenient, environmentally friendly properties and a therapeutic capacity for wound closure. Our results showed that ANFMs fabricated with high molecular weight polyvinyl alcohol (PVA) to form fibers during electrospinning had uniform fibrous architecture and a porous structure. Given the value of aloe gel in accelerating wound healing, liquid extracts from ANFMs significantly downregulated the expression of the pro-inflammatory genes, interleukin-6 (IL-6) and inducible nitric oxide synthase (iNOS), and markedly suppress the generation of reactive oxygen species (ROS) induced by lipopolysaccharide in RAW264.7 macrophages. These results indicated the excellent antioxidant and anti-inflammatory effects of ANFMs. After implantation into a mouse diabetic wound model for 12 days in situ, ANFMs notably expedited chronic wound healing via promoting angiogenesis and enhancing cell viability. Our ANFMs generated by handheld electrospinning in situ healed chronic wounds offer a convenient and promising alternative for patients to heal their own wounds under variable conditions.

Graphical abstract

1. Introduction

Diabetes mellitus is a metabolic disease that seriously jeopardizes human health and is the fourth leading cause of death after cardiovascular diseases, tumors, and AIDS [1]. Impaired wound healing is a typical complication of diabetes [2]. Chronic lesions that occur on feet can deteriorate to form ulcers that lead to high rates of amputation and morbidity [3,4]. Wound healing consists of overlapping and continuous phases comprising homeostasis and coagulation, inflammation, proliferation and migration, maturation, and remodeling [5]. Chronic inflammation [6,7] and excessive reactive oxygen species (ROS) [8] delay the healing of diabetic chronic wounds. Aloe vera is an evergreen perennial succulent plant that has been widely applied as a functional food, cosmetic, and pharmacological agent for millennia [9]. Over 75 active compounds in A. vera gel have been extracted and identified, including anthraquinones and naphthalenones, polysaccharides, proteins, enzymes, and organic acids [10]. Aloe vera gel has excellent antioxidant, anti-inflammatory, lipid- and glycemic-modulating effects [[10], [11], [12], [13]]. Aloe vera gel and extracts have been comprehensively explored to promote its clinical application to skin wounds in various forms, such as natural latex, oral medications, blended gels, and in creams [[14], [15], [16], [17]]. Among these, latex as a wound enclosure and gels do not permit aeration, and the indirect actions of oral medications impair their functions. Therefore, a novel strategy to develop an appropriate biomaterial based on the above issues is required for wound healing.

Electrospinning is a versatile and relatively simple way to form hydrogel-based nanofibers that have been applied to medicine and nanomaterials [18]. Such substrates usually have notable properties such as gas/fluid exchange, accelerated hemostasis, and they allow the introduction of bioactive compounds. However, conventional electrospinning units are ponderous, costly, and unsuitable for individual operation, as they comprise a pump or syringe, a high-voltage current supplier, and a metal plate collector [5]. Long et al. overcame these limitations by developing a portable battery-operated electrospinning apparatus (BOEA), which offers the advantages of individual accessibility and synthesis in situ because it is light and can be operated using a syringe pressed with a finger [19,20]. This device allows hemostasis as nanofibers can be directly electrospun onto wounds [20]. This, combined with its portability and effectiveness, the BOEA can be further exploited in diabetic wound healing by combining it with conventional herbal medicines, such as Aloe vera.

The current coronavirus disease 2019 (COVID-19) pandemic caused by SARS-CoV-2 substantially reduced outpatient clinic capacities and medical resources [21]. Hence, developing an effective strategy that can be applied at home by patients without physician supervision has become an increasingly acceptable and practical concept for widespread chronic wound healing. Polyvinyl alcohol (PVA) is biocompatible, biodegradable, non-toxic and can be easily electrospun [22]. Thus, we fabricated aloe-nanofiber membranes (ANFMs) containing high-molecular-weight PVA by electrospinning using the handheld BOEA. This strategy was economical, simple, environmentally friendly, and could heal chronic ulcers and lesions. We explored the anti-inflammatory and antioxidant properties of ANFMs in vitro, and verified their healing effects in diabetic mouse models of wound healing in vivo. Fig. 1 shows a schema of this process.

Fig. 1.

Schematic of ANFMs shows electrospinning using BOEA in situ and underlying mechanisms of wound healing mediated by ANFMs.

2. Materials and methods

2.1. Fabrication and characterization of ANFMs

Fig. 2A shows aloe gel extraction from the inner leaves of Aloe vera (Youpinkulasuo Horticulture, Guangdong, China). Briefly, the leaves were peeled and natural aloe gel obtained by twisting and squeezing with a medical gauze was passed through a filter (40 ​μm) to eliminate indiscerptible components. The filtrate was centrifuged at 931×g for 10 ​min, then supernatants were collected.

Fig. 2.

Fabrication and characterization of ANFMs.

(A) Preparation of PVA-based aloe solution. (B) Schema of ANFMs electrospun in situ using handheld device. (C) Morphology of PVA nanofibers and ANFMs. (D) Tensile mechanical properties of ANFMs. (E) FT-IR spectra of ANFM (9% PVA). ANFMs, aloe-nanofiber membranes; FT-IR, Fourier transform infrared; PVA, polyvinyl alcohol.

A PVA-aloe gel solution was obtained by directly dissolving 6%, 9%, and 12% w/v PVA (MW 80,000; Sigma Aldrich Corp., St. Louis, MO, USA) in the aloe gel extract. This was poured into a syringe and loaded into a portable BOEA (Qingdao Junada Technology Co., Ltd., Qingdao, China) for electrospinning as described by the manufacturer. The device was held to maintain 10 ​cm between the receiver (aluminum foil or skin wound) and the needle; and the syringe was gently pushed to extrude the solution for nanofiber formation. The ANFMs were harvested after 1 ​min of electrospinning, dried in vacuo for 2 weeks, then sputter-coated with gold for 60 ​s to increase conductivity. The morphology of samples was visualized using a type ∗/s-4800 scanning electron microscope (SEM) (Hitachi Ltd., Tokyo, Japan) at an acceleration voltage of 8–10 ​kV. We verified the components of the ANFMs by Fourier transform infrared (FT-IR) spectroscopy using a Nicolet iS20 spectrometer (Thermo Fisher Scientific Inc., Waltham, MA, USA) in the range of 600–4000 ​cm^-1. Tensile mechanical properties were measured using an Instron 5542 Advanced Material Testing System (Instron®, High Wycombe, UK) with a crosshead separation rate of 20 ​mm/min at room temperature.

2.2. Cell culture

We cultured RAW 264.7 macrophages (Chinese Academy of Science Shanghai Branch, China) in RPMI 1640 medium (Gibco Laboratories, Gaithersburg, MD, USA) at 37°C under a 5% CO₂ atmosphere. Fibroblasts were isolated from the dorsal skin of SD rats and cultured in Dulbecco’s modified Eagle Medium (DMEM; Gibco Laboratories) at 37°C under a 5% CO₂ atmosphere. Both media were supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin (both from Gibco Laboratories).

2.3. Biocompatibility of ANFMs

Fibroblasts (3 ​× ​10 [4]/well) suspended in DMEM were seeded into 96 well plates and incubated overnight, followed by starvation in serum-free DMEM for 24 ​h. We then suspended 0.5, 1, and 1.5 ​g of ANFMs (5%, 10%, and 15% w/v, respectively) in 10 ​mL of DMEM for 3 ​min, then removed undissolved ANFMs. The starved fibroblasts were cultured for 1 and 3 days in ANFM supernatants. The biocompatibility of the ANFMs was indirectly analyzed using CCK-8 assays. Briefly, the fibroblasts were incubated for 2 ​h with 10 ​μL of CCK-8 reagent in 90 ​μL of DMEM per well. Medium without cells was the blank control. Absorbance was measured at 450 ​nm using a microplate reader (Thermo Fisher Scientific Inc., Waltham, MA, USA). Cell viability was calculated as:

$$Cellviability(\%)=\frac{{OD}_{\mathit{exp}}-{OD}_{blank}}{{OD}_{control}-{OD}_{blank}}\times 100\%$$

where OD _exp, OD _blank and OD _control represent the optical density of the experimental, blank control, and control cells (cultured with DMEM), respectively, at 450 ​nm.

2.4. Hemolysis assay

Citrated mouse blood (2 ​mL) was diluted in 2.5 ​mL of phosphate buffered saline (PBS) and stored at 4°C, then ANFMs (10 ​mg) and diluted blood (0.2 ​mL) were incubated in 10 ​mL of PBS for 1 ​h at 37°C. The mixture was centrifuged at 3,000 ​rpm for 5 ​min, then absorbance was measured in supernatants at 545 ​nm using a microplate reader. The positive and negative controls comprised 0.2 ​mL of blood diluted in 10 ​mL PBS and deionized H₂O, respectively.

2.5. Antioxidant activity of ANFMs

We activated RAW 264.7 macrophages (1 ​× ​10 [6]/well) with lipopolysaccharide (LPS; 1 ​mg/mL) in 6-well plates, then incubated them with RPMI containing 5%, 10%, and 15% (w/v) aloe or various concentrations of ANFMs for 24 ​h. We then assessed the generation of reactive oxygen species (ROS) by staining the macrophages with 2′,7′-dichlorofluorescein diacetate (DCFH-DA) using ROS Assay Kits as described by the manufacturer (Beyotime Biotechnology, Shanghai, China). The intensity of fluorescence emitted by the macrophages was quantified using an inverted fluorescence microscope (Carl Zeiss AG., Oberkochen, Germany). The negative and positive controls were incubated with RPMI without or with LPS, respectively.

2.6. Anti-inflammatory activity of ANFMs

We activated RAW 264.7 macrophages (1 ​× ​10 [6]/well) with 1 ​mg/mL LPS in 6-well plates then incubated them for 24 ​h with RPMI containing 5%, 10%, and 15% (w/v) ANFMs. The mRNA expression of pro-inflammatory (IL-6, iNOS) and anti-inflammatory (CD206) markers was detected in macrophages using the quantitative real-time polymerase chain reaction (qRT-PCR). We reverse-transcribed mRNA isolated from the macrophages using EZ-press RNA Purification Kits (EZBioscience, Roseville, MN, USA) into cDNA using Color Reverse Transcription Kits (EZBioscience) as described by the manufacturer. Target DNA sequences were amplified by qRT-PCR using a QuantStudio 6 Flex Real-Time PCR instrument (Applied Biosystems, Waltham, MA, USA) with SYBR Green qPCR Master Mix (EZBioscience) and the primer sequences (Sangon Biotech Co., Ltd., Shanghai, China) listed in Table 1 . Relative expression was normalized to that of the housekeeping gene (18S).

Table 1.

RNA amplification forward (F) and reverse (R) primers.

Genes	Primer sequences (5′→3′)
IL-6	F: ATAGTCCTTCCTACCCCAATTTCC R: GATGAATTGGATGGTCTTGGTCC
CD206	F: AGACGAAATCCCTGCTACTG R: CACCCATTCGAAGGCATTC
iNOS	F: CAGAAGTGCAAAGTCTCAGACAT R: GTCATCTTGTATTGTTGGGCT
18S	F: CGGAACTGAGGCCATGATTAAG R: GTATCTGATCGTCTTCGAACCTCC

2.7. Animal tests in vivo

Twelve db/db mice (Nanjing University-Nanjing Biomedical Institute, Nanjing, China) were randomly assigned to control, PVA, and ANFM groups. The mice were anesthetized using intraperitoneal injection of pentobarbital sodium(50 ​mg/kg), hair on their backs was shaved, then a circular wound (6 ​mm diameter) was incised and covered in situ with handheld electrospun nanofiber membranes (∼ 50 ​mg) for ​∼ ​20 ​min (day 0). This procedure was repeated on the following day. The wounds were photographed 0, 3, 6, 9, and 12 days after the initial injury.

2.8. Histological analysis

Skin tissue samples excised 12 days after wounding were fixed in 4% paraformaldehyde for 24 ​h, embedded in paraffin, then sliced into 5 ​μm-thick sections. Epidermal regeneration was assessed by detecting collagen in sections stained with hematoxylin and eosin (HE) and Masson trichrome (both from Servicebio Wuhan, China) as described by the manufacturer. Histologically stained tissues were digitally recorded using a DS-Ri2 inverted microscope (Nikon Corp., Tokyo, Japan), and the thickness of the granulation tissue was measured using Image-Pro Plus 6 software (Media Cybernetics, Rockville, MD, USA).

2.9. Immunohistochemical analysis

We assessed neovascularization by immunohistochemical staining for CD31. Fixed sections were boiled in sodium citrate retrieval solution (pH 6.0) (Servicebio), then washed three times with PBS (pH 7.4). Non-specific antigen binding in sections was blocked by incubation for 30 ​min with 5% bovine serum albumin (BSA). The sections were then incubated overnight at 4°C with 1:400-diluted anti-CD31 mouse primary antibody (Servicebio) followed by 1:400-diluted HRP-goat anti-mouse secondary antibody (Servicebio) for 50 ​min at room temperature. The sections were then stained with 3, 3′-diaminobenzidine (DAB; Servicebio) reagent for 10 ​min, counterstained with HE for 3 ​min and assessed using a Nikon Eclipse E100 microscope (Nikon Corp).

2.10. Immunofluorescence analysis

Cell proliferation was visualized by immunofluorescence staining with Ki-67. Skin sections processed as described above for immunohistochemical staining were incubated with 1:500-diluted anti-Ki-67 rabbit primary antibody (Servicebio) overnight at 4°C followed by horseradish peroxidase (HRP)-goat anti-rabbit secondary antibody (diluted 1:200, Servicebio) for 50 ​min at room temperature. The sections were stained with DAB reagent and hematoxylin (as described for CD31), visualized using a PANNORAMIC MIDI II digital scanner (3DHISTECH Kft., Budapest, Hungary). The numbers of Ki-67-positive cells and neovascularization were determined from scanned images.

2.11. Statistical analyses

All values are presented as means ​± ​standard deviation (SD). Data were analyzed by one-sample and other t-tests using SPSS Statistics 26 (IBM Corp., Armonk, NY, USA). Values with P ​< ​0.05 were considered statistically significant.

3. Results and discussion

3.1. Fiber surface morphology of electrospun ANFMs in situ is uniform

Compared with traditional electrospun nanofibers, ANFMs electrospun in situ using BOEA have incomparable advantages in terms of production cost, convenience, security, and sustainable development. All raw materials for ANFMs are economical and easily obtained. Fig. 2A shows that aloe gel can be extracted from Aloe vera plants without complicated equipment. The ANFMs had an obvious membrane structure and excellent flexibility that allowed them to be easily peeled from the skin (Fig. 2B). Electrospinning in situ using the BOEA can be applied by patients at anytime, anywhere, without the need for medical supervision. Furthermore, the raw materials for ANFMs are not toxic, which ensures safe clinical applications.

The SEM findings showed that the fiber surface morphology of membranes containing 9% ANFMs was uniform (Fig. 2C), and similar to that of the 9% PVA nanofibers. The diameters of 9% PVA and ANFMs were both in the range of 1–2 ​μm, which was close to other PVA-based nanofibers [23]. In contrast, fibers in membranes containing 6% or 12% PVA had obvious desultory and conglutinant phenomena. We assessed the effects of the PVA content on the flexibility and strength of the ANFMs (Fig. 2D). We found that 9% ANFMs sustained the most tensile stress while maintaining excellent elasticity (∼ 280%), which exceeds that of human skin (60%‒75%) [23]. Based on these results, further analyses included nanofibers containing 9% PVA.

We verified the aloe gel content in the ANFMs by analyzing the Fourier transform infrared (FTIR) spectra of PVA and ANFMs containing 9% PVA (Fig. 2E). The FTIR spectra of 9% PVA and 9% ANFMs did not significantly differ at 3,299.76, 2,916, and 1,432 ​cm^-1 (O–H stretching, alkyl bending, and C–H stretching vibrations). However, significantly enhanced intensity ranging from 1,500 to 1,700 ​cm^-1 attributed to amino acid and phenyl groups of aloin A, aloin B, and aloe-emodin [24] indicated that the fabricated nanofibers contained aloe gel.

3.2. Cytocompatibility and hemocompatibility of ANFMs

The cytocompatibility of ANFMs was assessed as viability in vitro using an CCK-8. Cell viability in the control, PVA, and ANFMs groups did not significantly differ, suggesting that the nanofibers electrospun in situ were cytocompatible and would function well as wound dressings (Fig. 3 A). No integrated erythrocytes were evident in deionized H₂O in the hemocompatibility assessment of ANFMs (Fig. 3B). However, blood cells in the ANFM and PBS groups retained their normal morphology, indicating excellent ANFM hemocompatibility. These results were further verified in Fig. 3C, which shows obvious sediment of the PBS and ANFMs groups compared with the homogeneous solution in deionized H₂O. Furthermore, the absorption at 545 ​nm in the ANFM group was obviously lower than that in deionized H₂O (Fig. 3D), further implying excellent ANFM hemocompatibility.

Fig. 3.

Cytocompatibility and hemocompatibility of ANFMs.

(A) Relative cell viability of fibroblasts determined using CCK-8 assays (n ​= ​3). (B) Microscopy images of blood cells in PBS, ANFMs, and deionized H₂O. (C) Optical images of blood cell hemolysis. (D) Quantitation of blood cell hemolysis at 545 ​nm ^†p ​< ​0.001; n ​= ​3).

3.3. Antioxidant and anti-inflammatory activities of ANFMs

We analyzed the antioxidant activities of ANFMs using a DCFH-DA probe to detect endogenous ROS in RAW 264.7 ​cells. We found that LPS-stimulated ROS production in macrophages was remarkably reduced in the aloe groups compared with the positive control (Fig. 4 A and B). This indicated that the aloe gel had antioxidant activity. Furthermore, ANFMs in PVA nanofibers protected macrophages from damage caused by ROS (Fig. 4F and G), also indicating antioxidative activity.

Fig. 4.

Antioxidative and anti-inflammatory activities of ANFMs.

(A,B) Fluorescence images and intensity of aloe-induced ROS expression (∗p ​< ​0.05, ^†p ​< ​0.001; n ​= ​3). (C‒E) Gene expression of IL-6, iNOS, and CD206 in macrophages incubated with aloe in vitro (∗p ​< ​0.05; n ​= ​3). (F,G) Fluorescence images and intensity of ANFM-induced ROS expression (^†p ​< ​0.001; n ​= ​3). (H‒J) Gene expression of IL-6, iNOS, and CD206 in macrophages incubated with ANFMs (∗p ​< ​0.05; n ​= ​3).

Because ROS are associated with the NF-kB signaling pathway [25] that further stimulates the M1 polarization of macrophages [26], we assessed the anti-inflammatory properties of aloe-nanofibers using qRT-PCR. Fig. 4C‒E shows that aloe significantly decreased gene expression of the M1 phenotype markers iNOS and IL-6 and increased that of the M2 marker CD206. The ANFMs similarly induced macrophage differentiation into the M2 phenotype (Fig. 4H‒J). Macrophages adapt to various physiological states and maintain homeostasis, by polarization into classical pro-inflammatory (M1) and alternative anti-inflammatory (M2) phenotypes [27]. Our results implied that ANFMs have anti-inflammatory functions. Pro-inflammatory cytokines secreted by M1 macrophages, such as TNF-α, IL-6, and IL-12, contribute to inflammatory states and ROS [8]. Given the antioxidant and anti-inflammatory effects of ANFMs, we speculated that ANFMs electrospun in situ could prevent diabetic wounds from low-grade and self-perpetuating inflammatory states [7] and accelerate wound healing in vivo.

3.4. Epidermal and vascular regeneration promoted by ANFMs electrospun in situ

We investigated the effects of ANFMs electrospun in situ on promoting the healing of wounds (6-mm diameter) in the backs of db/db mouse models of diabetes (Fig. 5 A). The wounds were covered with ANFMs and photographed on days 0, 3, 6, 9, and 12 to evaluate healing. Fig. 5B shows significantly smaller wound areas in the ANFM, compared with the other groups after 12 days. The wound area rates were 114%, 28%, and 8% in the control, PVA, and ANFMs groups respectively, at day 12 (Fig. 5C). Masson and HE and staining results revealed significantly thicker granulation tissue in the ANFM group, again suggesting that ANFMs promoted healing (Fig. 5D and E). The anti-CD31 staining results revealed significantly increased capillary density in the chronic wound bed in the ANFM group, indicating vascular regeneration (Fig. 6 A, C). Epithelial regeneration was investigated by staining cell proliferation markers with immunofluorescent Ki-67 (Fig. 6B). The proportion of Ki-67-positive cells was notably higher in the ANFM, than in the other groups (Fig. 6D), further implying that ANFMs induced rapid cell regeneration [28].

Fig. 5.

Epidermal regeneration induced by ANFMs.

(A) Schema of healing process in diabetic mice in vivo. (B) Optical images of wound healing for 12 days. (C) Wound area rates (∗p ​< ​0.05, ^†p ​< ​0.01, ^‡p ​< ​0.001; n ​= ​4). (D) Masson and hematoxylin-eosin stained images of wounds at day 12. Arrows, granulation tissue width. (E) Actual width of granulation tissue (∗p ​< ​0.05; n ​= ​4).

Fig. 6.

Pro-angiogenesis and epidermal cell proliferation caused by ANFMs in vivo.

Diabetic wounds at day 12 stained immunohistochemically with anti-CD31 (A) and with (B) immunofluorescent anti-Ki-67. (C) Capillary density in images quantified by anti-CD31 staining. Rectangles, magnified areas. Arrows, blood vessels (∗p ​< ​0.05; n ​= ​4). D) Quantitation of Ki-67-positive cells. Rectangles, magnified areas. Arrows indicate Ki-67-positive cells (∗p ​< ​0.05; n ​= ​4).

In summary, ANFMs have distinct therapeutic properties in terms of improving wound closure, tissue regeneration, and angiogenesis. Angiogenesis and epithelial proliferation simultaneously intersect during wound healing. The neo-vasculature guides epithelial regeneration via perfusion nutrition and oxygen, and the epidermal closure of the wound allows neovascularization in a stable environment. In this context, insufficient angiogenesis in diabetic states prevents wound closure and healing [29]. Along with their antioxidant and anti-inflammatory properties, ANFMs accelerated wound healing in diabetic mice. These effects might be attributed to anthraquinones in aloe gel. Aloe-emodin and aloin are the main anthraquinones found in aloe. Aloe-emodin has antihyperglycemic, anti-inflammatory, and antioxidant properties. The anti-inflammatory properties of aloe-emodin resemble those of human oral epithelial cells [30] and it downregulates the LPS-induced inflammatory response in RAW264.7 ​cells [31]. Moreover, the antioxidant superoxide dismutase, immune-modulating polysaccharides, and antimicrobial acemannan in aloe gel are candidate factors that also promote wound healing [12]. Overall, the antioxidant and anti-inflammatory properties of ANFMs electrospun in situ conferred excellent epithelial and angiogenesis-improving effects on wound healing.

4. Conclusion

In summary, the morphology of ANFMs was superior to those of gauzes or hydrogels. This type of wound dressing offers the advantages of handheld electrospinning in situ, in which a porous fibrous membrane protects wounds from exogenous pathogens, and a herbal element significantly promotes healing via antioxidant and anti-inflammatory activities. Our findings in vitro and in vivo confirmed that ANFMs accelerated vascularization and promoted epithelial regeneration that consequently enhanced healing. Overall, handheld electrospun ANFMs offer an effective, available, economical, and environmentally friendly approach to practical individualized treatment for chronic ulcers and wounds in patients with diabetes. Furthermore, more bioactive materials with antioxidant and anti-inflammatory effects can be developed as individualized therapy for patients with diabetes using BOEA and electrospinning in situ.

CRediT authorship contribution statement

Chang Liu: Conceptualization, Methodology, Investigation, Writing – original draft, Writing – review & editing. Yun Wang: Methodology, Investigation, Writing – original draft. Pei Wang: Methodology, Investigation, Writing – original draft. Yan Gong: Writing – review & editing. Bingcheng Yi: Methodology, Formal analysis. Jing Ruan: Methodology, Formal analysis. Xiansong Wang: Conceptualization, Supervision, Funding acquisition.

Declaration of competing interest

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