The following reagents were purchased from various vendors: vanadium pentoxide (V2O5), oxalic acid dihydrate (C2H2O42H2O), isopropanol, cobalt nitrate hexahydrate (Co (NO3)26H2O), methanol, dimethyl imidazole (C4H6N2, Hmim), hydrochloric acid (HCl), ethanol (Sinopharm Chemical Reagent Co., Ltd., China), 3,3′,5,5′-tetramethylbenzidine (TMB, Shanghai Aladdin Biotechnology Co., Ltd., China), tryptone (LP0024, OXOID, USA), yeast extract (LP0021, OXOID, USA), agarose (Bacto agar, BD, USA), sodium chloride (NaCl, Guoyao Group Chemicals Co., Ltd., China), low growth factor matrix gel (Corning, USA), Tissue Protein Extraction Reagent (T-PER, Saimofeishier Technology Co., Ltd., China), Halt Protease and Phosphatase Inhibitor Cocktail (100× , Saimofeishier Technology), BCA Protein Concentration Determination Kit (Enhanced, Shanghai Biyuntian Biotechnology Co., Ltd., China), PVDF Transfer Membrane (Millipore, USA), ECL DualVue WB Marker (GE Healthcare, USA), SuperSignal West Dura Extended Duration Substrate (Semerfeld Technology Co., Ltd., China), Rg1 and Lig (Dingrui Chemical Co., Ltd., Shanghai, China), CCK-8 kit (Shanghai Biyuntian Biotechnology), 4% paraformaldehyde (Wuhan Boshide Bioengineering Co., Ltd., China), calcium fluorescent probes (Fluo A4, S1060, Shanghai Beyotime Biotechnology Co., Ltd., China), potassium chloride (KCl, Shanghai Macklin Biochemical Technology Co., Ltd., China), GelMA (GM-60, Suzhou Yongquan Intelligent Manufacturing Co., Ltd., China), Rat SDF-1α (CXCL12, Peprotech, USA), and Rat SDF-1 ELISA Kit (Shanghai Enzyme-linked Biotechnology Co., Ltd., China).

Synthesis of hollow VO2 and VO2@ZIF-67

VO2 was synthesized according to a hydrothermal method, as follows: 1.5 g V2O5 and 1.0 g dihydrate acetic acid were weighed in deionized water, continuously stirred at 70 °C for 60 min, and then added with isopropanol for 30 min at room temperature followed by reaction in a poly-tetrachloroethylene-lined reactor (100.0 mL) at 200 °C for 12 h. The black precipitate was centrifuged, collected, washed three times with ethanol, vacuum dried at 60 °C, and the resulting particles were collected to obtain the hollow VO2. The metal–organic framework material ZIF-67 was grown in situ on the prepared VO2 surface by co-precipitation. Briefly, 50.0 mg of VO2, 290.0 mg of cobalt nitrate hexahydrate, and methanol was added in a water bath for 10 min of ultrasound. The mixed solution was added to 16.5 mg/mL 1-methylimidazole methanol solution and stirred continuously for 20 min. After standing for 24 h, the black precipitate was centrifuged, collected, washed three times with ethanol, and dried in a vacuum oven at 60 °C. After drying, the particles were collected to obtain the composite VO2@ ZIF-67 (VZ).

Characterization of VO2 and VO2@ZIF-67

The morphologies and structures of VO2 and VZ were described using scanning electron microscopy (SEM, SU8010, Hitachi, Japan) and transmission electron microscopy (JEM-2100F, JEOL, Japan). The images were processed using ImageJ software to obtain the particle size distributions of VO2 and VZ. The nitrogen adsorption and desorption curves of VO2 and VZ were obtained using an automatic specific surface area and micropore analyzer. The elemental distribution of the VZ was analyzed by transmission electron microscopy (TEM) with an attachment to the spectrometer.

Preparation of gel/VZ

We dissolved 50.0 mg Lithium acylphosphinate salt (LAP) in 20.0 mL of ultrapure water to prepare a 2.5 mg/mL LAP aqueous solution. GelMA [33, 62] was dissolved in LAP aqueous solution in a water bath at 37 °C to prepare 10% (w/v) GelMA solution. After GelMA was completely dissolved, VZ was added and vortexed 1 min prior to exposure to 405 nm light to prepare 200 μg/mL Gel/VZ, which was stored in a refrigerator at 4 °C.

Rheological analysis of Gel/VZ

Rheological measurements of the ASD were performed with a rheometer (Anton Paar) using three different methods: (1) The strain amplitude sweep tests of GelMA and Gel/VZ were conducted at a fixed angular frequency (1 rad/s) with γ = 0.1%–300%; (2) The self-restoring abilities of GelMA and Gel/VZ were investigated by step frequency scanning at 37 °C with a fixed frequency of 1 rad/s. Amplitude oscillatory strains were alternated between little strain (γ = 1%) to greater strain (γ = 150%), each lasting for 60 s; and (3) The Frequency scanning of GelMA and Gel/VZ was carried out at a fixed strain (γ = 1%) in the range of ω = 0.1−100 rad/s.

pH-responsive Co2+ release of ASD

For pH responsive Co2+ release, Gel/VZ containing 400 μg/mL VZ was prepared and immersed in 5 mL solution with a pH of 7.4, 6.8, or 5.4 in a rotary shaker at 100 rpm, 37 °C. We supplemented 0.5 mL of the solution with fresh solution (0.5 mL) at predetermined time intervals. The amount of Co2+ released was determined using inductively coupled plasma mass spectrometry (NexION 300X, PerkinElmer). For pH responsive angiogenesis, 80 μL growth factor-reduced Matrigel basement membrane matrix (Corning) was added to individual wells of 48-well plates and allowed to polymerize at 37 °C for 30 min. After treatment with Gel/VZ at pH 7.4, 6.8, or 5.4 and PBS for 24 h, HUVECs were seeded onto solidified Matrigel at a density of 2 × 104 cells/well. The enclosed vessel networks were photographed under a microscope after 12 h incubation at 37 °C. The obtained images were analyzed using the Angiogenesis Analyzer in ImageJ for the length of tubules and the number of reticular structures.

Drug release profiles from ASD

We added 200 ng CXCL12, 2 mg Rg1, and 2 mg Lig to 600 μL Gel/VZ gel containing 200 μg/mL VZ, mixed the solution, and placed it in a square mold under 405 nm light for 1 min to prepare the ASD. The ASD was placed in a dialysis bag (COMW = 8000–14,000 KDa) with 5 mL PBS and then soaked in 15 mL PBS in a rotary shaker at 180 rpm. Then, 1 mL of supernatant was sampled for drug content determination at a predetermined time point, and 1 mL fresh PBS was added after each sampling. CXCL12 concentration was determined using the ELISA kit, and Rg1 and Lig concentrations were determined by HPLC under the conditions described in the Additional file 1.

Isolation and culture of BMSCs

BMSCs were isolated and cultured as previously described [63]. SD rats were supplied by Shanghai Laboratory Animal Co. (SLAC), Ltd., China. All experimental procedures were performed in accordance with the Zhejiang University guidelines for the welfare of experimental animals. Briefly, rat femurs were excised from the epiphysis and bone marrow was flushed out using a syringe with Dulbecco’s modified Eagle’s medium (DMEM, Gibco BRL) supplemented with 10% fetal bovine serum (FBS, Gibco BRL), l-glutamine, penicillin (50 U/mL), and streptomycin (50 U/mL). The cell suspension was placed in a 25 cm2 tissue culture flask (Corning) and cultured at 37 °C in 5% CO2. Subconfluent first passage cells were detached from the flask with 0.25% trypsin–EDTA for 2 min at 37 °C. The second- to fifth-generation BMSCs were used in subsequent experiments.

Cell recruitment assay

BMSC migration was tested using an RTCA DP instrument (ACEA Biosciences Inc.). Firstly, Gel/VZ-C and Gel/VZ-CLR were prepared with a VZ concentration of 200 μg/mL, CXCL12 concentration of 0.83 ng/μL, and Rg1 and Lig concentrations of 0.12 mg/mL. Then, 20 μL of Gel/VZ-C, Gel/VZ-CLR, or PBS was added to the bottom chamber of modified 16-well plates (E-Plate 16, ACEA Biosciences Inc.) and cross-linked by 405 nm light for 1 min. Subsequently, 145 μL of serum-free cell culture medium was added into the bottom chamber, the top chamber of the E-Plate 16 was filled with serum-free medium, and the membrane was hydrated and preincubated in the CO2 incubator (HF90, Health Force) at 37 °C for 1 h before obtaining a background measurement. After this incubation period, the BMSC suspension was seeded into the top chamber at 5 × 104 cells in 100 µL. The E-Plate 16 was assembled by placing the top chamber onto the bottom chamber and then placed in the RTCA DP station to automatically monitor the impedance value every 5 min for 48 h, which was expressed as a cell index value. All data were recorded using RTCA software.

BMSCs proliferation stimulated by multiple drugs

BMSCs were seeded in 96-well plates at a density of 5 × 103 cells/well and incubated in 5% CO2. After the cells adhered to the well, the culture medium was replaced, and PBS, Rg1, Lig, Rg1 + Lig, and VZ + Rg1 + Lig were added (the concentrations of Rg1, Lig, and VZ were 15 μg/mL, respectively). After 24 h of culturing in a CO2 incubator, the culture medium was removed according to the CCK-8 kit instructions. The culture medium was gently washed twice with sterile PBS; 100 μL complete medium and 10 μL working solution were added to each well. After incubation for 1 h, the absorbance of the supernatant was measured at 450 nm using a microplate reader.

SA-β-gal staining

Fifth-generation BMSCs were seeded onto 6-well plates at a density of 1 × 105 cells/well. After treatment with PBS, Rg1, Lig, Rg1 + Lig, or VZ + Rg1 + Lig were added (the concentrations of Rg1, Lig, and VZ were 15 μg/mL, respectively) for 24 h. The BMSCs were then stained with SA-β-gal staining solution (Beyotime) and incubated overnight at 37 °C. Images were captured using an inverted microscope (Nikon, Tokyo, Japan), and the number of positive cells was calculated using ImageJ software.

Neural differentiation of BMSCs in vitro

BMSCs were seeded onto confocal culture dishes at a density of 4 × 104 or 2 × 104 cells/dish, and incubated overnight in the CO2 incubator at 37 °C. PBS, Rg1, Lig, Rg1 + Lig, or VZ + Rg1 + Lig were added (the concentrations of Rg1, Lig, and VZ were 15 μg/mL, respectively) every two days. After being cultured for 7 d, the cells were washed with PBS, fixed in 4% paraformaldehyde (Solarbio Life Science) for 30 min, permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) for 20 min, and blocked with 10% goat serum (Boster Biological Technology) for 30 min. The samples were then incubated overnight at 4 °C with anti-nestin antibody (Omnimab) and anti-β3-tubulin antibody (Cell Signaling Technology), and detection was achieved by subsequent incubation with FITC-conjugated goat anti-mouse IgG H&L (Beyotime) and CY3-conjugated goat anti-rabbit IgG H&L (Boster Biological Technology) for 1 h at 37 °C, followed by DAPI staining and imaging under a confocal fluorescence microscope (BX61, Olympus).

Calcium imaging assay

After treatment with PBS or VZ-LR and culturing for 7 d, the changes in Ca2+ were measured using the fluorescent Ca2+ indicator Fluo-4 AM (Beyotime) by confocal microscopy, as previously described. Briefly, cells were washed twice with Hank’s Balanced Salt Solution (HBSS) twice and loaded with 2 μM Fluo-4 AM for 30 min in the dark. The cells were rinsed with HBSS twice and incubated for another 20 min at 37 °C to ensure that Fluo-4 AM had completely transformed into Fluo-4. Images were captured using a laser scanning confocal microscope (Nikon A1R). Fluo-4 AM was excited at a wavelength of 488 nm. After the fluorescent signal stabilized (F0), 4.1 mg/mL KCl was added to excite the cells, and the excitation (Ft) was recorded in real-time for 3 min with 10-s intervals. The changes in Ca2+ were reflected by relative fluorescence calculated as ΔF = Ft − F0.

Full-thickness wound model construction

We purchased 56 male SD rats, each weighing 140–160 g, from the SLAC. All animals were maintained under constant conditions (temperature = 25 ± 1 °C), with free access to standard chow and drinking water. All animal experimental procedures were performed in accordance with the guidelines and protocols of the Animal Experimental Ethics Committee of the Zhejiang University. The animals were anesthetized with an intraperitoneal injection of 3% sodium pentobarbital (30 mg/kg). Full-thickness excision wounds were made symmetrically (1.5 × 1.5 cm) by scalpel excision on the depilated back of each rat. Rats were randomly divided into eight experimental groups (n = 7 per group): blank, Gel/VZ (VZ composite with GelMA hydrogel and blank vector), CLR (CXCL12, Lig, and Rg1 mixed), Gel/VZ-LR (VZ composite with GelMA hydrogel loaded with Lig and Rg1), Gel/VZ-C (VZ composite with GelMA hydrogel loaded with CXCL12), Gel/VZ-CL (VZ composite with GelMA hydrogel loaded with CXCL12 and Lig), Gel/VZ-CR (VZ composite with GelMA hydrogel loaded with CXCL12 and Rg1), and Gel/VZ-CLR (VZ composite with GelMA hydrogel loaded with CXCL12, Lig, and Rg1). After treatment, all groups were dressed with transparent tegaderm to prevent infection and wound rehydration. All groups were treated every 2 d post-surgery for a total of six times. Wound healing was recorded by taking a picture every 2 d and measuring the wound area. The weights of the rats were also recorded. The wound-healing rate was calculated by dividing the difference between the area on day 0 and day n by the area on day 0.

Histological analysis

Histological analysis was performed on healed skin tissues and organs 17 days after wound treatment. Samples were fixed in 4% buffered paraformaldehyde, dehydrated, and then embedded in paraffin or OCT compound for slice preparation. The sample sections (5-μm thick) were stained with Masson’s trichrome staining (Keygen Biotech) according to the manufacturer’s protocol. The stained skin sections were observed using a laser scanning confocal microscope (VS120, Olympus). Organs including the heart, liver, spleen, lung, and kidney were extracted and cut into smaller sections, fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned into 5-μm-thick slices. The organ sections were stained with H&E (Keygen Biotech) and visualized using laser scanning confocal microscopy for the histological study of toxicity.

Immunofluorescent staining

Wound tissues were sampled on days 3 and 17 after treatment, embedded in an optimal cutting temperature compound, frozen, and sliced into 10-µm-thick sections at − 22 °C. Sections were then treated with primary antibodies against rabbit anti-CD31 (Abcam) overnight at 4 °C, followed by a 50 min treatment with goat anti-rabbit IgG secondary antibody (Abcam) at 37 °C and 10 min treatment with 4′,6-diamidino-2-phenylindole (DAPI). The stained slides were observed under an Olympus VS200 fluorescence microscope (Japan). To visualize the migration of BMSCs and regenerated nerves in vivo, tissue sections on day 3 and 17 were stained with antibodies against CD90 (ProteinTech), nestin (ProteinTech), and β3-tubulin (Cell Signaling Technology). CD90, nestin, and β3-tubulin signals were visualized using FITC- and CY3-conjugated secondary antibodies (Thermo Pierce). Nuclei were stained with DAPI. Using a laser scanning confocal microscope (VS200, Olympus), images of the sections were obtained for three randomly selected areas for the quantification of fluorescence intensity. All images were post-processed and quantified using ImageJ software.

Western blot analysis

Samples were obtained from the rat wound area and homogenized using T-PER tissue protein extraction reagent (Thermo Fisher Scientific). The protein concentration was determined using a bicinchoninic acid protein assay (Beyotime). Western blot analysis was performed using 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis. After the proteins were transferred onto PVDF membranes (Millipore), they were probed with antibodies against p-JAK2 (Cell Signaling Technology), JAK2 (Cell Signaling Technology), p-STAT3 (Cell Signaling Technology), STAT3 (Cell Signaling Technology), PTEN (Cell Signaling Technology), and GAPDH(Abcam) by incubation at 4 °C overnight. After washing with T-TBS, the membranes were incubated with the corresponding secondary antibodies (Thermo Pierce) for 1 h at room temperature. The blots were developed using SuperSignal West Dura Extended Duration Substrate (Thermo Pierce) and recorded on X-ray film (Fuji super RX); the bands were quantified using Image J.

Statistical analysis

All data were analyzed by one-way analysis of variance and expressed as the mean ± standard deviation (SD). Student’s t-test was used to evaluate statistical significance, with p < 0.05 considered statistically significant.