4D-printed hydrogel with precise regulation of viability and environment of stem cells for diabetic skin wound healing

Abstract

Diabetic wounds not only suffer from vascular and nerve damage, but also face the severe challenge of impaired stem cell activity. In recent years, although traditional tissue engineering strategies provide exogenous stem cells for the healing of diabetic wounds, they have not reversed the dilemma of stem cell proliferation and differentiation in a high-glucose environment. In this work, piRNA-hsa-32182 was first demonstrated to be highly expressed in diabetic wounds and significantly inhibit the differentiation and migration of adipose-derived mesenchymal stem cells (ADMSCs). Accordingly, a 4D-printed tissue engineering hydrogel and piRNA-hsa-32182 antagomir were prepared to precisely modulate the survival microenvironment of ADMSCs and accelerate diabetic wound healing. 4D printed tissue engineering hydrogels provided a highly ordered microenvironment for ADMSCs through internal space homogenization, thereby effectively improving the loading rate and survival rate of stem cells. In addition, piRNA-hsa-32182 antagomir effectively enhanced the migration of ADMSCs, thereby increasing the deposition and maturation of collagen on the wound and promoting angiogenesis. In summary, this study significantly improved the microenvironment of wounds through the synergy of bio-intelligent printing technology and gene expression regulation, providing a new clinical paradigm for the treatment of diabetic wounds.

1. Introduction

Diabetes is a common chronic metabolic disease characterized by systemic hyperglycemia and a variety of derivative diseases caused by vasculopathy and neuropathy, including diabetic wounds [[1], [2], [3]]. According to statistics, approximately 25 % of diabetic patients suffer from diabetic wounds, which can easily progress to gangrene, amputation, and even death [4]. Consequently, these ulcers remain a significant challenge in the clinical management of diabetic complications [5]. Studies have indicated that continuously elevated hyperglycemia significantly impair the formation of new blood vessels, resulting in local tissue hypoxia, arterial angiopathy, and lower extremity neuropathy, which collectively contribute to the development of diabetic wounds through chronic wound erosion [1,6,7]. Nowadays, clinical treatment strategy of diabetic wounds primarily rely on conservative methods like wound debridement and antibiotic therapy, which exhibit acceptably infections control but unsatisfactory effect in regulation of neovascularization disorders or persistent ulceration [8]. Therefore, the repair of diabetic wounds requires the development of microenvironment-regulation strategy to reverse the dilemma of cell regeneration in diabetic wounds.

In recent years, stem cell depletion has been recognized as an important cause of diabetic neovascularization. Therefore, stem cell therapy is expected to be developed as a strategy for achieving complete healing of diabetic wounds. [9]. Adipose-derived mesenchymal stem cells (ADMSCs), which are easily obtained from bone marrow, adipose tissue, synovial fluid, and other tissues of adults [10,11], can regulate immune response, promote angiogenesis, stimulate cell survival, proliferation and specific cell differentiation, regulate wound microenvironment by secreting anti-inflammatory cytokines and growth factors [[12], [13], [14]]. However, there are two major challenges in the clinical application of ADMSCs for diabetic wound repair. First of all, the existing strategies of ADMSCs treatment, whether direct injection or application of conventional tissue engineering scaffolds, provide unsatisfactory ADMSCs retention and wound healing. [15]. Due to the lack of precise microspace control, hydrogels prepared by traditional methods exhibit non-uniform pore size distribution and poor mechanical support, thus are unsuccessful in the support of an ideal survival microenvironment for stem cells as well as unsatisfactory cell translocation. [16]. Secondly, the microenvironment of diabetic wounds, in which the enhanced AGEs-RAGE binding triggers oxidative stress and inflammation, thus significantly impairing stem cell function and wound healing [17,18]. Therefore, the high retention rate of stem cells at the wound site and the reversal of microenvironment interference are two key targets for the improvement of current stem cell therapy for diabetic wounds. In addition, recent studies have further revealed the crucial role played by immune cells in wound repair, particularly T cells, which participate in coordinating inflammatory responses and tissue regeneration at the wound site [19]. The activation and migration of T cells has been reported to be regulated by Wnt signaling pathway, which could influence the balance between inflammation and tissue remodeling during cutaneous wound healing [20].

Currently, three-dimensional (3D) printing technology and four-dimensional (4D) printing technology have been proved to be a promising strategy for precise regulation of the living space of stem cells [21,22]. In previous research, we have successfully developed a 4D bioprinting technology, which utilized temperature-sensitive biodegradable hydrogel and a 3D bioprinting strategy to automatically print hydrogels to the desired shape and pore size according to design specifications, showing great application potential in the biomedical field [23]. The above 4D biopprinting technology is expected to produce hydrogel carriers with multi-scale porous structure to change shape over time [24], providing a microenvironment that is conducive to the nutritional uptake, growth, proliferation and metabolism of stem cells [25]. In addition, the abnormal expression and the regulation of diabetic wound microenvironment of various micro RNAs have attracted more and more attention. Piwi-interacting RNAs (piRNAs), small non-coding RNAs first identified in germ cells, have been implicated in cardiovascular diseases, cancer, neurodegenerative disorders, and diabetes complications through various molecular mechanisms and cellular pathways [[26], [27], [28], [29], [30], [31], [32]]. In a rat model of diabetes, silencing of the Piwil2 or Piwil4 gene resulted in impaired insulin secretion, indicating the involvement of piRNA in the development of type 2 diabetes [33]. Additionally, transcriptome analysis of human skin tissue revealed elevated levels of piRNA during skin healing, suggesting its regulation in wound healing [33,34]. Therefore, the combination of 4D printing technology and piRNA expression regulation is expected to provide comprehensive technical support for diabetic wound healing through precise regulation of stem cell and improvement of microenvironment.

In this study, high-throughput sequencing was firstly used to screen for piRNAs that are abnormally expressed in diabetic wounds. Consequently, piRNA-hsa-32182 was identified to play an important role in inhibiting the migration of ADMSCs at DFU wounds. Accordingly, a chitosan-based thermosensitive hydrogel carrier with uniform pore size and adjustable shape was developed using 4D bio-printing technology. Further, a 4D printed hydrogel carrier combined with piRNA-hsa-32182 inhibitor was used to load ADMSCs to complete the repair treatment of diabetic wounds. Specifically, the piRNA-hsa-32182 inhibitor and 4D-printed tissue engineering hydrogel were combined to precisely modulate the survival microenvironment of ADMSCs in diabetic wounds. On the one hand, 4D-CTH not only offers a highly uniformly porous structure and biomimetic mechanical properties to improve stem cell loading and survival rates, but also presents thermosensitive property to achieve more precise adhesion to the skin wound surface. On the other hand, the participation of piRNA-32182 antagomir plays a key role in the regulation of glucose metabolism in wound tissues, which promotes the migration of ADMSCs, the deposition and maturation of collagen, as well as the angiogenesis of diabetic wounds. In summary, this study significantly improves the microenvironment of diabetic wounds through simultaneous intervention of stem cell spatial environment and differentiation potential induction, thus providing a new clinical paradigm for new treatment strategies for diabetic wounds (see Scheme 1).

Scheme 1.

Preparation and the therapeutic mechanism of 4D-CTH + ADMSCs + piRNA antagomir for diabetic skin wound healing. (a) Procedure for the preparation of 4D-CTH + ADMSCs; (b) Procedure for treatment with 4D-CTH + ADMSCs + piRNA antagomir of skin damage in diabetic mice; (c) Mechanisms by which 4D-CTH + ADMSCs + piRNA antagomir to achieve cell proliferation, cell migration and angiogenesis promotion and reverse high glucose inhibition. Firstly, increased collagen deposition facilitates the reconstruction of the extracellular matrix. Secondly, the high expression of CD31 indicates enhanced angiogenesis. Thirdly, piRNA antagomir promotes epidermal growth by increasing the proliferation and migration of HaCaT cells.

2. Materials and methods

2.1. Clinical sample

The study protocol was approved by the Ethics Committee of the affiliated Hospital of Qingdao University (Shandong, China) (QYFY WZLL 27880). The ulcerated skin of eight patients with diabetic foot ulcers and the upper eyelid skin of five patients undergoing double eyelid surgery were collected from the affiliated Hospital of Qingdao University in 2021. All of the samples were collected during the operation, and then were immediately frozen in liquid nitrogen and stored in a cryogenic refrigerator at −80 °C for an extended period.

2.2. Sequencing and screening of piRNAs

Samples of ulcerated skin from patients with diabetic foot ulcers, and normal skin from patients undergoing double eyelid surgery were sent to Shu Pu (Shanghai, China) Biotechnologies LLC for high-throughput sequencing. The process of high-throughput sequencing was divided into five parts. 1) RNA sample quality control: The purity and concentration of total RNA samples were determined with NanoDrop ND-1000. 2) Sequencing Library Preparation: Total RNA of each sample was sequentially ligated to 3′and 5'small RNA adapters. The cDNA was then synthesized and amplified usingllumina's proprietary RT primers and amplification primers. Subsequently, 142–153 bp PCR amplified fragments were extractedand purified from the PAGE gel. And finally, the completed librarieswere quantified by Agilent 2100 Bioanalyzer. 3) Libraries denaturation and dilution: The libraries were denatured and diluted to a loading volume of 1.3 ml and loading concentration of 1.8 pM. 4) sequencing on illumina NextSeq 500: Diluted libraries were loaded onto reagent cartridge and forwarded to sequencing run on illumina NextSeq 500 system using NextSeq 500/550 V2 kit (#FC-404-2005, illumina), according to the manufacturer's instructions. 5) Data Collection and Analysis.

The sequencing results were analyzed through blast on the National Center for Biotechnology Information website. piRNAs were screened according to two principles including “unique gene locus” and “length between 21 and 35 nucleotides”. Then, the relative expression of piRNAs in the diabetes group and control group and the fold-change value were controlled to be more than 100 and more than 1.5, respectively. Finally, the piRNAs that met the above requirements were screened.

2.3. Cell culture and high-glucose treatment

Human keratinocytes (HaCaT cells) were purchased from Procell Life Science & Technology Co., Ltd (Wuhan, China). Cell culture requires 15 % fetal bovine serum (Biological Industries, Israel), 1 % penicillin–streptomycin solution (Dalian Meilun Biotechnology Co., Ltd., China), and Minimum Essential Medium (MEM) (Wuhan Procell Life Technology Co., Ltd., China). The cells were cultured in a cell incubator at 37 °C with 99 % relative humidity and 5 % CO₂. The HaCaT cells were treated with 10, 20, 30, and 40 mM of D-(+)-glucose (Sigma, USA) for 24 h, followed by a Scratch Assay.

2.4. RNA extraction and quantitative real time polymerase chain reaction analysis (qRT-PCR)

The total RNA in skin tissues (ulcerated skin from patients with diabetic foot ulcers, and normal skin from patients undergoing double eyelid surgery) and HaCaT cells (treated with 10, 20, 30, and 40 mM of D-(+)-glucose) were extracted using TRIzol (Takara, Japan) reagent. The concentration of RNA in each sample was measured using a BioDrop spectrophotometer (Bio-Tek, USA). Then, the total RNA was reverse-transcribed into cDNA with a special kit for stem-loop miRNA cDNA synthesis (Nanjing Novozan Biotechnology Co., Ltd., China). The piRNAs cDNA were reverse-transcribed using a T100™ thermal cycler(Bio-rad, USA). The level of piRNAs was detected by SYBR Green I chimeric fluorescence method, qRT-PCR analysis was carried out using the CFX Connect™ real-time system (Bio-rad, USA). The relative expression level of the gene was evaluated using the 2-ΔΔCT method. U6 was used as the internal control. The primers used in this study are listed in Supplement Table 1.

2.5. Oligonucleotide transfection

HaCaT cells were seeded in a six-well plate and transfected when the cell fusion rate reached 40 %–60 %. With Lipofectamine™ 3000 as a carrier, piRNA-hsa-32182 agomir group, piRNA-hsa-32182 antagomir group, agomir normal control (NC) group and antagomir normal control group, were transfected into HaCaT cells (Supplement Table 2). Specifically, the agomir and antagomir of piRNA-hsa-32182 were both designed by Shanghai GenePharma Co.,Ltd (Shanghai, China). The transfection efficiency was detected by qRT-PCR to compare the expression level of piRNA-hsa-32182 agomir group, piRNA-hsa-32182 antagomir group, agomir NC group and antagomir NC group.

2.6. Scratch assay

HaCaT cells were seeded in a six-well plate. When the cell cell fusion rate reached 100 %, the monolayer cells were cut vertically using a 200 μL pipette tip, and the representative images of cell migration were acquired 0 and 24 h after scratching. The shortened distance of the middle scratch was measured and normalized to the control for 0 h, and the results are presented as the relative migration rate.

2.7. Preparation of the 4D-CTH scaffold material

Chitosan (CTS, degree of deacetylation ≥95 %, viscosity: 100–200 mPa s) was purchased from Shanghai Maclean Biochemical Co., Ltd. (Shanghai, China). Carboxymethyl chitosan (CMCTS, degree of deacetylation: 96.44 %, 134 kDa) was prepared according to the previous description [35]. β-glycerophosphate disodium was purchased from Shanghai Soraby Biotechnology Co., Ltd. Under aseptic conditions, 0.1–0.3 g of CTS was dissolved in 4.5 mL acetic acid solution (0.1 mM) and stirred for 10 h. Then, 0.2g CMCTS was dissolved in 4.5 mL double distilled water (4.4 %); β-glycerophosphate disodium was also dissolved in 2 mL double distilled water (6 %). Adding β-glycerophosphate disodium solution to CMCTS solution (drop by drop addition; operating in 4 °C). After that, CTS solution was mixed with this solution, and the combined material was named CTH. The CTH was printed into a homogeneous carrier by using a 4D bio-printer (Bio-Architect® SR, Regenovo, China) stacking layer by layer at a low temperature (0–4 °C). Then, the materials are stored by undergoing vacuum freeze-drying.

2.8. Characterization of the 4D-CTH scaffold

Electron micrography: the 4D-CTH sheets underwent vacuum freeze-drying, and then, the products were sprayed with gold and observed and photographed using a scanning electron microscope (Tescan, Brno, Czech Republic); Swelling rate assay: The freeze-dried scaffold materials were placed in double-distilled water until equilibrium swelling was reached and the weight no longer increased. Then, the swelling ratio of the scaffold materials was analyzed. Storage modulus and loss modulus: The CTH, 4D-CTH (printed into 10-mm-diameter and 1-mm-thick slices), were placed on the support plate of the rheometer. The storage modulus and loss modulus of the gel slices were measured at a shear frequency of 0.1–10 Hz and strain of 1 %. Viscosity assay: The CTH, 4D-CTH (printed into 10-mm-diameter and 1-mm-thick slices), and adipose tissue from the back of a pig, were placed on the support plate of the rheometer.

The attenuated total reflectance Fourier transform infrared spectroscopy (FTIR) spectra were recorded on a Nicolet™ iS50 FTIR spectrometer (Thermo Scientific, Waltham, MA, USA), covering a frequency range from 400 to 4000 cm⁻¹. The 4D-CTH samples were frozen and lyophilized in a freeze dryer, and then, all of the studied samples were prepared in KBr. Data analysis and drawings were carried out using OriginPro 9.1.

2.9. Extraction and culture of ADMSCs

All animal procedures were conformed to the Guide for the Care and Use of Laboratory Animals and performed following the guidelines and the protocol approved by the application of the experimental animal ethical project of Qingdao Agricultural University (Approval No.20220068). ADMSCs were acquired from C57BL/6J mice aged 6–8 weeks (SPF, (Beijing) Biotechnology Co., Ltd., China). After the mice were executed, the inguinal fat was acquired and immersed in 1000 U/mL of penicillin-streptomycin for 10 min. The blood vessels and connective tissues without adipose tissues were separated under the Asana microscope (Olympus, USA) and then were digested in 10 mL 0.1 % collagenase I (Solarbio, Beijing, China) and were placed in a shaker (150 rpm, 37 °C) for 1 h. The digestion was terminated by the addition of 10 mL complete medium (10 % fetal calf serum, 100 U/mL penicillin-streptomycin, 2 mM L-glutamine) and centrifuging at 1500 rpm for 10 min. After the supernatant was discarded, the precipitant was eluted with normal saline and centrifuged at 1500 rpm for 10 min. This process was repeated for twice. The cells were centrifuged (1500 rpm for 10 min) after cell sieve filtration (100 μm) and re-suspended in a complete culture medium (10 % fetal calf serum, 100 U/mL penicillin-streptomycin, 2 mM L-glutamine). The cells were cultured in a special basic culture medium for Mesenchymal Stem Cells (MSCs) (Dakewe, Shenzhen, China), which contained 5 % serum substitute (EliteCell, USA) and 1 % penicillin-streptomycin. ADMSCs were cultured in an incubator at 37 °C.

2.10. Identification of mouse adipose-derived mesenchymal stem cells

ADMSCs during the second generation were identified following three methods: observation of cell morphology, identification of cell surface marker molecules by flow cytometry, and identification of cell multidirectional differentiation ability. When the cell fusion rate reached 60 %–80 %, the cells were digested with pancreatic enzyme (Dalian Meilun Biotechnology Co., Ltd., China) for 1.5 min, centrifuged at 1000 rpm for 5 min, and washed with Phosphate Buffered Saline (PBS) (Dalian Meilun Biotechnology Co., Ltd., China). Then the cells were sealed by cell staining buffer (Elabscience, China) containing Bovine Serum Albumin (BSA) for 20 min. A total of seven eppendorf tubes were prepared (included CD44, CD90, CD45, CD11B, IgG2a, IgG2b, and a control tubes without stain solution), and then washed twice with cell staining buffer. Fluorescein Isothiocyanate (FITC)-labeled CD44, CD90, CD45, and CD11b and isotype controls IgG2a and IgG2b were added to each remaining tube. The cells with above solutions were incubated at 4 °C for 30 min without light, and then were washed twice with cell staining buffer. Then the cells were suspended with 200 μL of staining buffer and detected using flow cytometry (Beckman Coulter, Inc.). All the antibodies were purchased from Elabscience Biotechnology Co., Ltd. (Wuhan, China).

2.11. Evaluation of multidirectional differentiation ability of ADMSCs

First, a six-well plate was prepared. Then, the 0.1 % gelatin was placed at the bottom and dried. The second-generation ADMSCs were inoculated in the coated six-well plate at a cell density of 2 × 10⁴ cells/cm², with 2 mL of complete medium (10 % fetal calf serum, 100 U/mL penicillin-streptomycin, 2 mM L-glutamine) was added to each well. When the cell density reached 100 %, the primary culture medium was absorbed. According to the instruction of adipogenic differentiation kits (OriCell, Guangzhou, China), the Lipogenic inducer A was added. After 3days, the Lipogenic inducer B was used. After 24 h, the Lipogenic inducer B was replaced by Lipogenic inducer A. This cycle was repeated 3 times. The induction was stopped until there were a large number of fat droplets of suitable size in the cells. The oil red O staining was performed to detect the production of lipids.

ADMSCs (2 × 10⁴ cells/cm²) were inoculated in the coated six-well plate. When the cell fusion reached 60 %, the complete medium (10 % fetal calf serum, 100 U/mL penicillin-streptomycin, 2 mM L-glutamine) was replaced with a MSCs medium (Fuyuanbio, Shanghai, China) for osteogenic induction. The fresh medium for osteogenic induction (Fuyuanbio, Shanghai, China) was replaced every 3 days, and the cells were cultured for 2–4 weeks. Then the state of the cells was observed, and the induction was terminated when significant calcium deposition of the cells was observed. Alizarin red staining (1 ml) was added for 15 min to detect the formation of calcium nodules.

2.12. Cytocompatibility detection of 4D-CTH

ADMSCs were used to evaluate the effect of 4D-CTH on cell migration (scratch assay). The ADMSCs culture medium was replaced with 4D-CTH extract for the experiment group, which was a cell culture medium soaked for 7days with 4D-CTH and containing 0.2 g/mL dried scafold powder. Cells cultured in medium alone served as the normal control. ADMSCs (2 × 10⁴ cells/cm²) were seeded in six-well plates for 24h. After the cells were fully grown, the cells were vertically scratched with a 200-μL pipette tip, rinsed twice with PBS, replaced with serum-free medium, and stained with Calcein AM (Dalian Meilun Biotechnology Co., Ltd., China) at 0, 12, and 24 h after scratching. The images of scratch assay were acquired by using an inverted fluorescence microscope (Olympus, Japan). The scratch width was quantified by using ImageJ software to calculate the wound-healing rate.

Live/dead cell staining (Calcein AM/PI) was used to evaluate the effect of 4D-CTH on cell viability. Living/Dead Cell Double-staining Kits (AM/PI) were used for quantitative analysis of live/dead cells. First, ADMSCs were seeded in 12-well plates at a density of 1 × 10⁵/well for 24 h. The medium was then replaced with fresh medium in the control group, with medium containing 4D-CTH (0.2 g/mL dried scafold powder) in the 4D-CTH group. According to the guiding protocol of the AM/PI kit, after 24, 48, and 72 h of co-culture with 4D-CTH, the cells were gently washed with PBS 2–3 times. Then, 4 μL AM and 4 μL PI were mixed in a centrifuge tube and 4 mL of PBS solution was added to obtain a 1000 × dye solution. Finally, the cells were incubated with the dye solution at room temperature in the dark for 15 min. The staining of the cells was observed using an inverted fluorescence microscope. The fluorescence intensity of live and dead cells was calculated by using ImageJ software.

2.13. In vivo wound healing of 4D-CTH+ADMSCs+piRNA-hsa-32182 antagomir

The animal experimental protocol was approved by the Animal Ethics Committee of Qingdao university (No.20220421C572020220811016). C57BL/6J mice aged 6 weeks were purchased from SPF Biotechnology Co., Ltd., (Beijing, China), and adaptive feeding was performed at the Laboratory Animal Center of Qingdao University for 1 week. All mice were fed with High-Fat diet for 5 weeks. Diabetes was induced by intraperitoneal injection of streptozotocin (STZ) at a concentration of 50 mg/kg (Solarbio, Beijing, China) for 5 consecutive days. A week later, tail vein blood was collected, and blood glucose concentration was detected using a portable blood glucose meter. The diagnostic criterion of diabetic mice was a blood glucose level >13.9 mmol/L [36].

Eight weeks after the STZ injection, the model was successfully created. The mice were anaesthetized by isoflurane inhalation (1.5 % in O2), and the mice were shaved and deeply depilated using depilating cream. The skin of the mice was disinfected with iodophor. Full-thickness skin wounds were created on the back of the mice by 5-mm skin punch biopsy (Miltex, USA) in a biosafety cabinet to establish the diabetic mouse skin wound model. The experimental mice were divided into five groups with four mice in each group: for control group, only 50 μL of PBS was applied to the skin wound of each diabetic mouse; for the 4D-CTH group, only the 4D-CTH vector was used to repair the skin wound model of diabetic mice; for the ADMSCs group, the skin wound model of diabetic mice was repaired with 50 μL of ADMSCs (1 × 10⁶ cells); for the piRNA-hsa-32182 antagomir group, the skin wound model of diabetic mice was repaired using 50 μL of piRNA-hsa-32182 antagomir (20 μM); for the 4D-CTH + ADMSCs + piRNA-hsa-32182 antagomir group, piRNA-hsa-32182 antagomir and 4D-CTH loaded with ADMSCs both play important role in the repair of the skin wound of diabetic mice. Images of wound healing were acquired at 0, 4, 7, and 14 days after the establishment of the wound model. ImageJ software was used to analyze the wound area, and the percentage of the wound area at different time points was obtained by comparing the wound area with the original wound area. Wound area (%) = (current wound area/initial wound area) × 100 %.

2.14. Histological analysis

Mice were anaesthetized on the 7th and 14th days of wound healing, and wound tissues were collected for histological analysis. Skin tissues were fixed with 10 % paraformaldehyde (Solarbio, Beijing, China) and embedded in paraffin, and 3-μm paraffin sections were cut using a paraffin slicing machine (Leica, Germany). After the sections were dewaxed and covered with water, they were stained with hematoxylin and eosin (H&E). Finally, the slices were sealed with dehydrated, transparent, and neutral balsam (Solarbio, Beijing, China). After the sections were allowed to dry, they were examined by using an inverted microscope.

2.15. Masson's trichrome staining and sirius red staining

To evaluate the collagen production in the wound, the skin tissues on the 7th and 14th days were stained with Masson's trichrome. Paraffin sections were deparaffinized and hydrated. Masson's trichrome staining solution was purchased from Fuzhou Maixin Biotechnology Development Co., Ltd (China). Masson's compound staining solution A was dropped into the sections to ensure that the staining solution covered the tissues, and the staining solution was incubated for 5 min and washed with distilled water. After that, the phosphomolybdic acid solution (solution B) was dropped and incubated for 5 min. Then, aniline blue solution (solution C) was added and incubated for 5 min. Finally, the acetic acid differentiation solution (solution D) was added and incubated for 30 s. The final procedure was followed by dehydrated, transparent, and neutral balsam sealing. The staining was observed by using an inverted microscope (Olympus Corporation IX73).

To further evaluate the formation of collagen I and collagen III, paraffin sections were stained with Sirius red. The Sirius red staining kit was purchased from G-Clone (Beijing) Biotechnology Co., Ltd (China). The paraffin sections were deparaffinized and hydrated, stained with Weigert's iron hematoxylin staining solution for 15 min, rinsed with tap water for 5 min, and then rinsed with distilled water once. Then, the slices were stained with Sirius red staining solution for 1 h, washed twice with acidified water working solution, and finally dehydrated, transparent, and sealed with neutral balsam. The staining was also observed by using an inverted microscope.

2.16. Immunofluorescence staining

The wound skin tissues were acquired on the 7th and 14th day. Frozen sections with a diameter of 6 μm were cut using a frozen microtome, were sfixed in 4 % paraformaldehyde for 15 min, and then washed three times with PBS for 5 min each time. The cells were then blocked with blocking buffer (5 % BSA+0.3 % Triton X-100) at room temperature for 1 h. Then the blocking buffer was discarded and incubated with Ki67 (Cell Signaling Technology, 1:150, Alexa Fluor 488 labeled, USA) antibody overnight at 4 °C. The tissues were incubated with α-smooth muscle actin (α-SMA) (Cell Signaling Technology, 1:200, USA) antibody overnight at 4 °C and then with Cy3-labeled goat anti-rabbit IgG secondary antibody (Boster Biological Technology Co., Ltd., 1:100, Wuhan, China) for 2 h at room temperature. Finally, the tissues were sealed with an anti-fluorescence attenuation tablet containing DAPI (Dalian Meilun Biotechnology Co., Ltd., China). Images were captured by using an inverted fluorescence microscope.

2.17. Immunohistochemical staining

After the paraffin sections were deparaffinized and hydrated, sodium citrate antigen repair solution (Solarbio, Beijing, China) was added for microwave antigen repair, and an appropriate amount of endogenous peroxidase blocker (ZSGB-BIO, Beijing, China) was added for incubation at room temperature for 10 min. The sections were incubated with CD31 (Cell Signaling Technology, 1:100, USA) primary antibody overnight at 4 °C. Tissues were labeled with an enzyme-conjugated sheep anti-rabbit IgG polymer (ZSGB-BIO, Beijing, China) and incubated at 37 °C for 20 min, then appropriate DAB chromogenic solution was added and incubated at room temperature for 3 min. After that, hydrochloric acid ethanol differentiation solution (Beyotime, Shanghai, China) was added to differentiate and wash for reverse blue. Finally, the tissues were decolorized by alcohol, rendered transparent by xylene, and sealed with neutral balsam. Images were captured using an inverted microscope and finally analyzed by ImageJ software.

2.18. Spatial transcriptomics

The spatial transcriptome profiling was performed using 10x Visium platform from 10x Genomics on unfixed on 5 μm paraffin sections. Samples were processed as follows: Detection of RNA Quality in Section Samples (DV200 > 50 %), tissue adhesion detection; after quality inspection; FFPE tissue sections are placed on expression chips, H&E staining is used to observe whether the morphology of tissues meets the formal experimental requirements; Experimental sample patches meeting the requirements are selected, probe hybridization, probe connection, tissue permeabilization, probe extension, cDNA amplification, library establishment and sequencing are carried out; After obtaining sequencing data, the barcoded library is mapped back to specific data points on the capture area. The original sequencing data was compared to the reference genome (GSE142471) using data from 10x Genomics. The Cell Ranger software is processed for single-cell gene counting. The SCTransform algorithm in Seurat v3.0 Biotuctor package 88 was used to normalize the single cell count data.

2.19. Statistical analysis

All of the data are presented as mean ± Standard Error of Mean (SEM). Unpaired Student's t-test was used for comparison between two groups, and one-way Analysis of Variance (ANOVA) was used for comparison between multiple groups. Statistical analysis was performed using GraphPad Prism 8.0 software (GraphPad Software, Inc., USA), and p < 0.05 was considered to be statistically significant (∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001).

3. Results

3.1. The highly expressed piRNA-hsa-32182 in DFU inhibited HaCaT cells migration

piRNA is an important factor regulating the progression of diabetes disease, which has been confirmed by a large number of studies, but the function and mechanism of piRNA in the occurrence and development of diabetes-derived diseases (especially diabetic foot ulcers, DFU) have not received sufficient attention and analysis. Therefore, in this work, skin samples from normal skin and affected areas of DFU were first compared and analyzed by high-throughput sequencing to clarify the association of piRNA with the development of DFU disease and to obtain abnormally expressed piRNA with specific sequences (Fig. 1A). Compared with normal skin samples, a total of 63 up-regulated piRNAs (red marks) and 64 down-regulated piRNAs (blue marks) were found in foot ulcer skin samples (Fig. 1B). Further, the above abnormal expression piRNA was screened according to the following three criteria: (1) the expression level of piRNA in both groups was greater than 100; (2) The expression ratio of piRNA between the two groups was greater than 1.5; (3) piRNA was differentially expressed in HaCaT cells cultured with high glucose as in human samples (Fig. 1A). In the end, only piRNA-hsa-32182 was confirmed to be the targeting piRNA which met all of these criteria. qRT-PCR further confirmed the increased expression of piRNA-hsa-32182 in both diabetic foot ulcers and HaCaT cells cultured with high glucose (Fig. 1C and D). Moreover, the expression level of piRNA-hsa-32182 in HaCaT cells was positively correlated with glucose-induced concentration (5–40 mM) (Fig. 1D). Wounds in high glucose environment showed decreased healing compared with normal wounds. Therefore, high expression of piRNA-hsa-32182 could be associated with impaired cell migration and wound healing in diabetic tissues. To confirm the biological role of piRNA-hsa-32182 in diabetic wound healing, cell models with high and low expression of piRNA-hsa-32182 were established by transfection of piRNA-hsa-32182 agomir and piRNA-hsa-32182 antagomir. qRT-PCR confirmed the expression status of piRNA-hsa-32182 in overexpression (OE) group and knockdown (KD) group, respectively, demonstrating the successful establishment of cell models with high-expression or low-expression of piRNA-hsa-32182 (Fig. 1E and F). Further, wound healing result of HaCaT cells proved the negative correlation of expression of piRNA-hsa-32182 with the migration of cells (Fig. 1G and H). The above results confirmed that abnormally high expression of piRNA-hsa-32182 in diabetic wounds could significantly inhibit cell migration by disrupting glucose metabolism, leading to delayed wound healing. Therefore, the development and application of piRNA-hsa-32182 inhibitor is expected to promote diabetic wound repair.

Fig. 1.

piR-hsa-32182 is upregulated in diabetic wounds and inhibits the migration of HaCaT cells. A, Schematic diagram of the screening piRNAs. B, Scatter plot of high-throughput sequencing. C, qRT-PCR analysis of piRNA-hsa-32182 expression levels in diabetic and normal skin tissues. D, qRT-PCR analysis of piRNA-hsa-32182 expression levels in HaCaT cells treated with different concentrations of glucose. E and F, Transfection efficiency of agomir and antagomir transfected in HaCaT cells (abbreviation: OE: overexpression; OE NC: overexpression control; KD: knockdown; KD NC: knockdown control). G, Quantitative statistical data of scratch assay. H, Representative picture of a scratch assay taken by an inverted microscope, in which HaCaT cells were treated with 40 mM glucose and the width of scratch was measured after overexpression and knockdown of piRNA-hsa-32182; Scale bar:200 μm; Data are presented as mean ± SD (n = 3 independent experiments). ∗p < 0.05,∗∗p < 0.01 versus NC group. Unpaired Student's t-test was used for comparison between two groups, and one-way Analysis of Variance (ANOVA) was used for comparison between multiple groups.

3.2. The extraction and identification of ADMSCs

Tissue engineering technology based on stem cell replenishment has been recognized as a promising clinical strategy for the clinical treatment of diabetic wounds. ADMSCs, a type of adult stem cells available from patients, have been shown to regulate the wound microenvironment through the secretion of anti-inflammatory cytokines and growth factors and participate in the diabetic wound healing, thus showing great clinical translational potential. In this study, ADMSCs were extracted from the adipose tissue of the groin of mice according to the operation steps in Fig. 2A. In addition, the morphology, markers, and characteristic behaviors of the extracted cells were identified to confirm the successful extraction of ADMSCs. The primary cultured cells grew adherently and presented a long fusiform shape, exhibiting typical ADMSCs characteristics (Fig. 2C). Positive markers of ADMSCs, CD90 and CD44, positive marker of leukocyte, CD45, and positive marker of macrophage, CD11b were respectively used to identify ADMSCs. Flow cytometry confirmed the positive expression of CD90 and CD44 in the extracted cells and the negative expression of CD45 and CD11b, which not only identified the extracted cells to be ADMSCs but also excluded the possibility of white blood cells and macrophages (Fig. 2Bⅱ and 2Bⅴ). In addition, the negative expression of IgG2a and IgG2b in the control group and the homologous control group indicated consistent immunophenotype of the extracted cells with that of ADMSCs (Fig. 2Bⅰ, 2Bⅵ and 2Bⅶ). Oil red O staining showed good lipid forming ability of the extracted cells, in which there were a large number of red spherical lipid droplets with strong refraction (Fig. 2D). Furthermore, alizarin red staining results confirmed the presence of red calcium nodules in the cells, demonstrating the satisfactory osteogenic ability of the extracted cells (Fig. 2E). Therefore, oil red O staining and alizarin red staining confirmed the lipid-forming and bone-forming differentiation capabilities of the extracted cells, respectively, determining the extracted cells with stem cell characteristic. In summary, the above results confirmed the successful extraction of ADMSCs as an ideal seed source for subsequent stem cell tissue engineering therapy. To investigate whether piRNA-hsa-32182 influenced the migration and differentiation of ADMSCs, scratch assays were performed to assess ADMSCs migration following piRNA-hsa-32182 overexpression and knockdown, and the results demonstrated that elevated piRNA-hsa-32182 expression inhibited ADMSC migration (Fig. 2F). Furthermore, Alizarin Red and Oil Red O staining confirmed that high piRNA-hsa-32182 expression suppressed the osteogenic and adipogenic differentiation of ADMSCs (Fig. 2G). In conclusion, increased piRNA-hsa-32182 expression was successfully proved to negatively regulate both the migration and differentiation of ADMSCs.

Fig. 2.

Morphology and identification of mouse ADMSCs. A, Schematic diagram of the extracting ADMSCs. B, Surface marker molecules of the extracted ADMSCs were detected by flow cytometry; (ⅰ) blank control group (without fluorescent primary antibody) was negative; (ⅱ) ADMSCs positive marker CD90 positive; (ⅲ) ADMSCs positive marker CD44 positive; (ⅳ) ADMSCs marker CD45 negative (leukocyte marker); (ⅴ) ADMSCs negative marker CD11b negative (macrophage marker); (ⅵ) ADMSCs isotype control IgG2a negative (excluding non-specific binding of Fc segment); (ⅶ) ADMSCs isotype control IgG2b negative (excluding non-specific binding of Fc segment); C, Morphological observation of mouse ADMSCs. D, Oil red O was used for cytochemical staining to analyze the representative images of adipocyte differentiation of ADMSCs. E, Alizarin red was used to analyze the representative images of osteocyte differentiation of ADMSCs. D (ⅰ) and E (ⅰ) are photographs taken before staining. Scale bar:100 μm (left and middle); Scale bar:50 μm (right). F, Representative picture of a scratch assay, in which piRNA-hsa-32182 were overexpressed or knockdown in ADMSCs; Scale bar:200 μm. G, Oil red O and Alizarin red were used to analyze the representative images of ADMSCs after overexpression and knockdown of piRNA-hsa-32182; Scale bar:50 μm (left); Scale bar:100 μm (right). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

3.3. The characterization of 4D-CTH

While ADMSCs, with high differentiation ability and low immunogenicity, have been identified as seed cells, the preparation of carrier materials used to support ADMSCs is also crucial for the successful repair of diabetic wounds. In this work, chitosan, carboxymethyl chitosan and sodium β-glycerophosphate solution were mixed and prepared into chitosan-based thermosensitive hydrogel (CTH) and 4D-printing hydrogel according to the schematic diagram in Fig. 3A. The prepared CTH showed fluidity at 4 °C and a semi-solid state when the temperature rose to 37 °C, proving thermosensitive property of CTH hydrogel (Fig. 3B). According to the FTIR of CTH hydrogel, the absorption peaks at 1573 cm⁻¹ and 1409 cm⁻¹ represented the asymmetric and symmetrical stretching vibration of COO-, and the absorption peaks at 3237 cm⁻¹ represented the stretching vibration of -NH₂ [23], which accords with the structural changes after gelling (Fig. 3C). According to the determined model parameters (x = 40 mm, y = 40 mm, z = 1.0 mm), the CTH hydrogel at 4 °C was extruded layer by layer by an extruded 3D bio-printer under gas pressure at −20 °C, and freeze-dried 4D-CTH with uniform pore size was finally prepared (Fig. 3D). The prepared 4D-CTH exhibited morphological changes when the temperature rose to 37 °C, forming different shapes based on the model (Fig. 3E). This suggested that 4D-CTH can provided more satisfactory adaption to wounds with various morphologies. Intriguingly, 4D-CTH could get attached to the skin when joint bent, suggesting satisfactory joint-adaptable feature and adhesiveness of 4D-CTH (Fig. 3F). SEM result confirmed more regular structure of 4D-CTH than CTH hydrogel, which provided more uniform microenvironment for the proliferation of ADMSCs (Fig. 3G and H). The variation of the storage modulus (G′) and loss modulus (G″) with frequency and shear rate-viscosity of 4D-CTH were measured by rheometer. According to frequency scanning results, the storage modulus and loss modulus of 4D-CTH increased with the increase of frequency and G′ was always greater than G″, indicating satisfactory hydrogel performance of CTH and 4D-CTH (Fig. 3I). In addition, when evaluating the viscosity change trend with shear rate, 4D-CTH showed a performance closer to that of natural adipose tissue than CTH, proving that 4D-CTH was more in line with the mechanical properties of natural skin tissue, thus providing a more satisfactory environment for stem cell proliferation (Fig. 3J). Therefore, in this study, 4D-CTH, with regular three-dimensional structure are expected to provide an ideal microenvironment for the nutrition, growth and metabolism of stem cells, thereby improving the diabetic wound healing by ADMSCs.

Fig. 3.

Synthesis and physical behavior of 4D-CTH. A, Schematic diagram of the 4D-CTH's production. B, The thermoreversible sol-gel transition of the CTH between 4 °C and 37 °C. C, Analysis of CTH by ATR-FTIR. D, The structure of 4D-CTH; Scale bar = 10 mm. E, Schematic diagram and representative picture of 4D-CTH after temperature change; Scale bar = 10 mm. F, The picture of adhesiveness at joints. G, Electron micrograph and pore width analysis of 4D-CTH (n = 3); Scale bar = 200 μm. H, Electron micrograph and pore width analysis of CTH (n = 3); Scale bar = 200 μm. I, Storage modulus and loss modulus of 4D-CTH in the frequency range of 1–100Hz. J, The viscosity analysis of CTH, 4D-CTH and adipose tissue.

3.4. ADMSCs loading and stem cell characterization maintenance of 4D-CTH

According to the schematic process shown in Fig. 4A, ADMSCs were loaded into the 4D-CTH carrier, and the distribution, proliferation and differentiation of ADMSCs in the 4D-CTH carrier were evaluated, respectively. AM/PI staining and CCK-8 assay results proved more uniform distribution and higher cellular activity of ADMSCs in 4D-CTH than in un-printed CTH hydrogel (Fig. 4B, Fig. S1A). In addition, after co-culture with 4D-CTH for 24 h, 48 h, and 72 h, ADMSCs maintained typical long spindle cell morphology, and the cell survival rate was significantly higher than ADMSCs in un-printed hydrogel (Fig. 4C, Fig. S1B). These findings indicated that 4D-CTH, with more regular microenvironment and mechanical properties close to those of natural tissue, thus being expected to provide an excellent growth environment for ADMSCs. In addition, wound healing results showed that the ADMSCs cultured on 4D-CTH had greater cell proliferation and migration ability than ADMSCs in un-printed hydrogel (Fig. 4D, Fig. S1C). And oil red O staining and alizarin red staining respectively showed stronger lipid differentiation ability and osteogenic differentiation ability of ADMSCs cultured in 4D-CTH than that of un-printed hydrogel (Fig. 4E and F). Studies have shown that stem cells are often prone to aggregation and agglomeration under the influence of gravity during in vitro culture, which not only limits the free growth of stem cells, but also may affect the quality and function [37]. The above results confirmed that, compared with un-printed hydrogels, 4D-CTH, as a three-dimensional scaffold material with highly ordered and controllable spatial structure, can provide a more favorable spatial environment for the growth and stem cell characteristic maintenance of ADMSCs. 4D-CTH not only effectively improved the load and retention of ADMSCs in wound site, but also maintains the activity, migration ability and bone fat formation ability of ADMSCs, thus showing great potential for improving the stem cell repair effect of diabetic wounds.

Fig. 4.

The cytocompatibility assays of 4D-CTH. A, Schematic diagram of the ADMSCs mounted on 4D-CTH. B, The live/dead cell staining (AM/PI) Images of ADMSCs in 4D-CTH; Scale bar:500 μm; G1: 4D-CTH group; G2: CTH group. C, Images of AM/PI taken under an inverted microscope after co-incubation with 4D-CTH for 24, 48, and 72 h; Scale bar:200 μm; G1: Control group; G2: CTH group; G3: 4D-CTH group. D, Images of wound healing assay taken at 12 h and 24 h after AM staining in control group and 4D-CTH group; Scale bar:500 μm; G1: Control group; G2: CTH group; G3: 4D-CTH group. E, Oil red O was used for cytochemical staining to analyze the representative images of adipocyte differentiation of ADMSCs; G1: 4D-CTH group; G2: CTH group. F, Alizarin red was used to analyze the representative images of osteocyte differentiation of ADMSCs; G1: 4D-CTH group; G2: CTH group. E (a) and F (a) are photographs taken before staining. Scale bar:100 μm (left and middle); Scale bar:50 μm (right). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

3.5. 4D-CTH loading ADMSCs and piRNA-hsa-32182 antagomir promoted diabetic wound healing

Previous analysis of this study (Fig. 1) confirmed that abnormally high expression of piRNA-hsa-32182 at the diabetic wounds significantly inhibited cell migration and wound healing, indicating the potential of piRNA-hsa-32182 antagomir in promote diabetic wound healing. In this study, 4D-CTH loaded with ADMSCs and piRNA-hsa-32182 antagomir were prepared and evaluated for promoting diabetic wound repair. According to the schematic process in Fig. 5A, after the diabetic mouse model was successfully established (Supplement Tables 3 and 4), the mice were randomly divided into the following groups: Control group (G1), ADMSCs group (G2), piRNA antagomir group (G3), CTH hydrogel + ADMSCs + piRNA antagomir group (G4), 4D-CTH + ADMSCs + piRNA antagomir group (G5). After 14 days of treatment, 4D-CTH + ADMSCs + piRNA antagomir group showed more significant wound healing than the control group and other gourps (Fig. 5B). It is worth noting that under the same ADMSCs and piRNA antagomir, 4D-CTH showed a more significant promotion on wound healing than unprinted CTH hydrogels, indicating the important role of orderly tissue microenvironment in stem cell migration. The wound areas in G1-G5 on the 4th day were 84.60 %, 66.98 %, 71.64 %, 82.88 % and 49.48 % of the original wound area respectively. And on the 14th day, the wound area of the 4D-CTH + ADMSCs + piRNA antagomir group was about 10 % of the initial area, showing most obvious effect on promoting wound healing (Fig. 5C and D). Furthermore, in order to verify the impact of different treatments on the functional repair of the skin tissue, the mechanical properties of the skin and the regeneration of nerves were evaluated at 14 days. The results showed that the tensile strength of the skin tissue treated by 4D-CTH + ADMSCs + piRNA antagomir was comparable to that of normal mouse skin. And the expression of PGP 9.5, a neuronal axon marker, was significantly higher in the 4D-CTH + ADMSCs + piRNA antagomir group than in the control group, indicating favorable recovery of neural regeneration in the repaired skin (Fig. S1D–F). These results confirmed the successful therapy of piRNA antagomir combined with 4D-CTH loaded with ADMSCs in promoting diabetic wound healing.

Fig. 5.

Effects of different treatments on diabetic wound healing in vivo. A, Schematic diagram of the animal experiment. B, Representative images of wound healing under different treatment conditions captured at 0, 4, 7, 14 of wound healing; G1: control group; G2: piRNA antagomir group; G3: ADMSCs group; G4: CTH + ADMSCs + piRNA antagomir; G5: 4D-CTH + ADMSCs + piRNA antagomir. C, Statistical analysis results of wound area at 0, 4, 7, and 14 days under different treatment conditions; D, The healing area of different groups. Data are presented as mean ± SD (n = 4 independent experiments). ∗p < 0.05,∗∗p < 0.01 versus control group. The ANOVA was used for comparison between multiple groups.

3.6. Histological evaluation of diabetic wound healing promoted by 4D-CTH loaded with ADMSCs and piRNA-hsa-32182 antagomir

Further, the diabetic wound healing by piRNA-hsa-32182 antagomir and 4D-CTH loaded with ADMSCs were evaluated through histological evaluation. Specifically, H&E staining and Sirius Red staining of wounds after 7 days of treatment were adopted to evaluate tissue morphology and collagen deposition, respectively (Fig. S2A and S2B). The H&E results showed that, at the 14th day, the epidermal thickening and wound healing in the 4D-CTH + ADMSCs + piRNA inhibitor group were more significant than those in the control group and other treatment groups (Fig. S2A, S3A and B). The 4D-CTH + ADMSCs + piRNA antagomir group exhibited a more mature skin structure on day 14 than the other groups, including further increased hair follicles (Fig. 6A, S3C). Therefore, H&E staining confirmed a synergistic effect of 4D-CTH, ADMSCs, and piRNA antagomir on epithelial cell regeneration and matrix remodeling in diabetic wound healing. In addition, collagen is the main component of extracellular matrix [38]. Masson's trichrome staining is a classic and reliable collagen staining technology that can stain collagen blue to observe the morphology and distribution of newly born collagen. On the 7th day of wound healing, collagen in the control group showed sparse and disordered distribution. Compared with the control group, collagen arrangement in the CTH + ADMSCs + piRNA antagomir and 4D-CTH + ADMSCs + piRNA antagomir groups was more order and dense (Fig. S2B). On the 14th day of wound healing, the advantage of 4D-CTH + ADMSCs + piRNA antagomir in promoting collagen deposition at wound site was more prominent, which confirmed the spatial regulation ability of 4D-CTH and the advantage caused by the combined action of ADMSCs + piRNA antagomir (Fig. 6B). In addition to the content of collagen is crucial for wound healing, the type of collagen also plays an important role in the regulation of wound healing. In the process of normal skin wound healing, immature type III collagen is deposited by myofibroblasts at the initial stage of healing [39]. According to Sirius Red staining (Fig. 6C–S2C), type I collagen deposition with strong orange or bright red color, type Ⅲ collagen deposition with green color were significantly higher in 4D-CTH + ADMSCs + piRNA antagomir group than that in control group and other treatment groups. Furthermore, the results of immunofluorescence staining showed the increased expression of type I and type III collagen and closer ratio of type III collagen to type I collagen to that of normal mouse skin in the 4D-CTH + ADMSCs + piRNA antagomir group (Fig. S3D and S3E). It was confirmed that 4D-CTH, ADMSCs and piRNA antagomir also had significant effects on the types and distribution of collagen in connective tissues, which contributed to the healing of diabetic wounds. The above results prove that the therapeutic strategy of 4D-CTH + ADMSCs + piRNA antagomir accelerates the healing of diabetic wounds by promoting the epithelialization process of skin tissue and the deposition and maturation of collagen, which not only benefits not only from the regulation of ADMSCs and piRNA antagomir but also from the tissue microenvironment closer to the human body provided by 4D-CTH.

Fig. 6.

Assessment of the degree of epithelialization and collagen production of the wound. A, Representative H&E staining images captured on the 14th day of wound healing; Scale bar:100 μm (top); Scale bar:50 μm (bottom); B, Masson's trichrome staining images captured on the 14th day of wound healing under different treatment conditions; Scale bar:100 μm (top); Scale bar:50 μm (bottom). C, Sirius red stained images captured under polarized light microscope on the 14th day of wound healing; Scale bar:100 μm (top); Scale bar:50 μm (bottom). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

3.7. 4D-CTH loaded with ADMSCs and piRNA antagomir promoted cell proliferation, cell migration and angiogenesis of diabetic wound

Clinically, after debridement and antibacterial treatment, diabetic wounds are still faced with severe healing challenges which is mainly due to the deterioration of the microenvironment, including inhibition of cell proliferation, migration and angiogenesis. Therefore, in order to explore the regulatory mechanism of the combined strategy of 4D-CTH + ADMSCs + piRNA antagomir on the microenvironment during the accelerated process of diabetic wound healing, the protein markers of epidermal cell proliferation (Ki67), migration (α-SMA) and angiogenesis (CD31) were respectively detected. Firstly, Ki67, a protein marker of cell proliferation, was selected to evaluate the cell proliferation capacity of diabetic wounds in each group. The results in Fig. S4A and Fig. S4B showed that compared with the control group, more Ki67 positive cells were found in wound subepidermal granulation tissue in CTH + ADMSCs + piRNA antagomir group and 4D-CTH + ADMSCs + piRNA antagomir group on the 7th day after treatment. And the 4D-CTH + ADMSCs + piRNA antagomir group was the most significant (p < 0.01). On the 14th day, Ki67 expression in 4D-CTH + ADMSCs + piRNA antagomir group was still significantly higher than that in other groups (Fig. 7A, 7D, S4C). In addition, abundant neovascularization can promote wound healing through adequate supply of oxygen and nutrients, thus up-regulation of angiogenesis marker CD31 is an important signal for improvement of macrovascular and microvascular diseases [40]. Immunohistochemical staining of CD31 confirmed that the neovascular density of all treatment groups on day 7 and day 14 was higher than that of control group (Fig. S5A and 7B). Moreover, the 4D-CTH + ADMSCs + piRNA antagomir group showed the highest neovascular density (p < 0.001) (Fig. S5B and 7E). Moreover, in order to evaluate the improvement of cell migration at diabetic wound site by treatment of 4D-CTH + ADMSCs + piRNA antagomir, the expression of α-SMA, which is an important positive regulatory marker of cell migration ability [41,42], was further evaluated. Immunohistochemical results showed that α-SMA expression in the 4D-CTH + ADMSCs + piRNA antagomir treatment group was significantly increased on day 7 (Fig. S6A and S6C) and day 14 (Fig. 7C and F, S6B) compared with other treatment groups and control group (p < 0.01). In summary, the expression evaluation of Ki67, CD31 and α-SMA confirmed that the 4D-CTH + ADMSCs + piRNA antagomir strategy effectively improved cell proliferation, angiogenesis and cell migration in diabetic wounds (Fig. 7G), which has important clinical significance for the comprehensive regulation and repair of the overall environment of diabetic wounds.

Fig. 7.

The situation of cell proliferation, angiogenesis, and cell migration in wound. A, Ki67 immunofluorescence staining to evaluate the effect of different treatments on cell proliferation in 14th day; Scale bar:100 μm. B, Representative images of CD31 immunohistochemical staining on the 14th day after different treatments in each group; Scale bar: 100 μm (top); Scale bar: 50 μm (bottom). C, α-SMA immunofluorescence assay to assess cell migration in the tissue in 14th day; Scale bar:200 μm. D, Quantitative analysis of Ki67 fluorescence intensity on the 14th day of wound healing. E, Relative average density quantitative analysis of CD31 immunohistochemical staining in skin tissue of each group on the 14th day. F, Quantitative analysis of α-SMA fluorescence intensity on the 14th day of wound healing. G, Mechanism diagram of wound healing. Data are presented as mean ± SD (n = 4 independent experiments). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, versus control group. The ANOVA was used for comparison between multiple groups.

3.8. 4D-CTH + ADMSCs + piRNA antagomir rugulated microenvioment of diabetic wound by activating the Wnt signaling pathway in T cells

Spatial transcriptomics technology provides the possibility to study single cells derived from skin tissue from a spatial perspective, which is expected to reveal the mechanism of action of the microenvironment around wound healing [36,37]. In this study, spatial transcriptomics was used to compare the effects of different treatments on wound healing in diabetic mice, including 4D-CTH + ADMSCs + piRNA antagomir group and control group. On day 14 after treatment, wound tissue samples from both groups were collected and given 10x Genomics spatial transcriptome sequencing (Fig. 8A). First, eight major cell types, including T cells, were successfully identified by combining the single-cell dataset GSE142471 (Fig. S7A). Subsequently, spatial transcriptome data provided the spatial distribution of different cell types in the 4D-CTH + ADMSCs + piRNA antagomir group and the control group (Fig. S7B). The accuracy of cell classification was further confirmed by correlation analysis of the ratio between cells (Fig. S7C and S7D). There was no significant difference in the number of epithelial cells and keratinocytes between the 4D-CTH + ADMSCs + piRNA antagomir group and the control group, which may be related to the completion of wound healing in the 4D-CTH + ADMSCs + piRNA antagomir group at day 14. However, after 14 days of treatment, the number of T cells in the 4D-CTH + ADMSCs + piRNA antagomir group was still significantly higher than that in the control group (Fig. 8B), indicating that the repair and regulation of the wound microenvironment by the immune system is still necessary and crucial in a certain stage after epithelialization. Immune cells play a crucial role in different stages of wound healing, among which regulatory T cells (Tregs) promote wound healing by releasing anti-inflammatory cytokines [43]. Although most studies have shown that killer T cells impede tissue healing, constant natural killer T cells can promote skin wound healing by inhibiting persistent neutrophil inflammatory responses. UMAP cell cluster analysis firstly divided the T cells of the two groups into three distinct subpopulations (Fig. S8A): T1, T2 and T3. Subsequent localization and quantitative analysis of the three cell subtypes at the spatial level (Fig. 8C) confirmed that 4D-CTH + ADMSCs + piRNA antagomir treatment significantly activated T1 cells at the wound site (Fig. S8B), suggesting the positive role of T1 cells in the wound healing process. Further, the analysis of gene expression in T1 cells showed that the TOP 5 genes whose expression significantly changed after 4D-CTH + ADMSCs + piRNA antagomir treatment were Ptpn3, Krt83, Dhcr24, Nsun5 and R3hdm2 (Fig. S8C–E). Among them, the expression level of Ptpn3 was the highest in the cells of T1 subtype, and there was a significant difference between them and the other two subgroups (Fig. 8D). The Ptpn3 gene has been shown to be responsible for encoding the protein tyrosine phosphatase non-receptor 3 (PTPN3), which regulates inflammatory responses and cell proliferation through dephosphorylation of a variety of signaling molecules, and thus plays an important role in wound healing [44]. In addition, PTPN3 is involved in regulating signaling pathways associated with diabetes complications, including insulin signaling and oxidative stress responses [45]. GO enrichment analysis further confirmed that T cells with high expression of Ptpn3 were closely related to cellular metabolism, suggesting that Ptpn3 could promote wound healing by regulating T metabolism status (Fig. 8E). KEGG enrichment analysis revealed that T cells with high expression of Ptpn3 were significantly correlated with WNT signaling pathway, which was further verified by GSEA (Fig. 8F). Activation of the Wnt pathway can be involved in inflammatory responses, cell proliferation, angiogenesis, hair follicle regeneration, and skin fibrosis during wound healing [46,47]. To validate the above results, a transwell assay was performed to evaluate the impact of Wnt signaling on the migratory ability of HaCaT cells (Fig. S9A). Subsequently, T cells were treated with the Wnt signal pathway inhibitor IWR-1, which provided strong inhibitory effects while maintaining approximately 80 % cell viability (Fig. S9B and S9C). The results showed significantly suppressed Wnt signaling in T cells as well as reduced expression of β-catenin. Moreover, inhibition of the Wnt pathway led to a significant reduction of the migration of HaCaT cells (Fig. S9D and S9E). Therefore, the regulation of T cells through Wnt signaling may represent a potential target for further investigation to optimize wound repair. However, the specific mechanisms by which the Wnt signaling pathway regulates T cell differentiation and activation and affects skin wound healing still require further research to be clarified. The above findings revealed that 4D-CTH + ADMSCs + piRNA antagomir could improve the diabetic immune microenvironment by up-regulating PTPN3-expressing T cells and Wnt signaling pathway in the later stage of wound healing.

Fig. 8.

The spatial transcriptome analysis of combination therapy groups and untreated groups. A, spatial transcriptome sequencing diagram. B, The histogram of cell type proportion. C, The SPOTlight analysis of T cell subsets. D, The violin map of gene expression. E, GO analysis of differential genes of T1. F, KEGG analysis of T1 differential genes. G, GSEA analysis of T1 differential genes.

4. Conclusion

In this work, high-throughput sequencing was first used to screen piRNA-hsa-32182 and validate its key regulatory role for migration and differentiation in diabetic wounds. Based on the exploration of abnormal piRNA expression in diabetic skin wounds, we developed a 4D printed tissue engineering treatment strategy loaded with mesenchymal stem cells and molecular drugs for diabetic skin wounds. The results showed that piRNA-hsa-32182 antagonist and 4D-printed hydrogel contributed to the proliferation and migration of ADMSCs, and significantly promoted collagen synthesis, neovasculation and activation of immune microenvironment in the wound microenvironment. This study provides microenvironment optimization for ADMSCs from molecular regulation and living space simulation to effectively promote the repair of diabetic wound. In conclusion, this work not only provides new insights into the treatment of diabetic wound, but also opens up new directions for the application of stem cell therapy in complex wound repair, demonstrating the broad prospects of tissue engineering technology in regenerative medicine.

CRediT authorship contribution statement

Junlin Lv: Writing – original draft, Methodology, Data curation. Meng Li: Writing – original draft, Validation, Data curation. Xiaomin Wang: Writing – review & editing, Investigation. Liang Zhang: Resources. Dan Han: Resources. Xiaodan Hao: Writing – review & editing, Resources. Xiao Xu: Validation, Investigation. Yiwei Xu: Investigation. Yan Tang: Investigation. Zhen Shang: Validation. Nailong Pan: Investigation. Xiaoying Kong: Writing – original draft, Supervision, Conceptualization. Wenhua Xu: Supervision, Funding acquisition, Conceptualization.

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.

Appendix B. Supplementary data

The following is the Supplementary data to this article.

Data availability

Data will be made available on request.