Twenty-four female New Zealand white rabbits (weighing 2,000 ± 250 g) were provided by the Animal Experimental Center of Wenzhou Medical University and housed with sufficient water and food in animal cages for 14 days before surgery. All experimental procedures were performed under the regulations of the committee of the Institutional Animal Care and the Ethics Committee of Wenzhou Medical University (wydw 2019-0949).

The isolation and culture procedures of DPSCs were performed as previously described (Luo et al., 2021). In brief, dental pulp tissues, which were collected from the tooth, were cut into small pieces and digested using the mixture of collagenase type I (Gibco, United States) and dispase (Sigma, Germany) for 30 min at 37°C in a 5% CO₂ cell incubator. Then, the pulp tissue was suspended and cultured using α-modified Eagle’s medium (α-MEM, Gibco, United States) supplemented with 20% fetal bovine serum (FBS, Gibco, United States), 100 μg/ml of streptomycin, and 100 U/mL of penicillin (Gibco, United States) at 37°C in a 5% CO₂ cell incubator. The culture medium was replaced every 3 days. MSC-like characteristics of DPSCs were identified by flow cytometry, immunofluorescence analysis, and multi-differentiation detection. Flow cytometry was executed using human primary antibodies, including CD19, CD166, CD73, CD90 (BD Pharmingen™, United States), CD34, and HLA-DR (BioLegend, United States), and the data were evaluated using a CytoFLEX flow cytometer (Beckman Coulter, California, United States). The immunophenotype of DPSCs was analyzed using primary antibodies against CD44 (1:400, Abcam, Cambridge, United Kingdom), and the immunofluorescence of DPSCs was observed using a fluorescence microscope (Eclipse 80i, Nikon, Japan). Based on the procedures described in our previous work (Albashari et al., 2020; Luo et al., 2021), the multipotent property of DPSCs was analyzed by osteogenic, adipogenic, and chondrogenic differentiation with alizarin red S staining, Oil Red O staining, and Alcian blue staining, respectively. Images were observed using a light microscope (TS100, Nikon).

The effects of DPSCs on murine macrophage RAW264.7 cells were evaluated by performing a macrophage-DPSCs coculture experiment as previously described (Németh et al., 2009; Chen et al., 2019). First, the mouse monocyte-macrophage cells (RAW 264.7, ATCC, United States) were cultured in 24-well plates at a concentration of 2 × 10⁵ cells/well. After incubation for 24 h, the supernatants were removed, and the cells were treated with lipopolysaccharide (LPS, Sigma; 1 μg/ml) for 24 h. Then, DPSCs (2 × 10⁴ cells/well), which were cultured in insert membranes of the transwell system (.4 μm pore size, Corning), were transferred to the wells. After incubation for another 24 h, the supernatants were collected and stored at −20°C for further use. The expression of tumor necrosis factor-alpha (TNF- α ) and interleukin-10 (IL-10) of challenged cells and supernatants was assessed by immunofluorescence staining and Enzyme-linked immunosorbent assay (ELISA) (R&D Systems), respectively.

To evaluate the angiogenic characteristics of DPSCs, following in vitro experiment was performed to assess the formation of capillary-like structures. 96-well plates were pre-coated with gelatin methacryloyl (GelMA) hydrogel as previously described (Chen et al., 2012). In brief, DPSCs and human umbilical vein endothelial cells (HUVECs, ATCC, United States) were plated onto hydrogel-treated 96-well plates and cultured with an endothelial cell growth medium-2™ (EGM-2™, Lonza Bioscience, Switzerland) supplemented with 2% FBS, .4% hFGF-B, .1% VEGF, .1% hEGF, .1% R3-IGF-1, .1% Heparin, .1% ascorbic acid, .1% gentamicin/amphotericin-B, and .04% hydrocortisone. The cultured medium was changed every 2 days. After differentiation for 7 days, the cells were fixed with 4% PFA and treated with Phalloidin-TRITC (1:200, Solarbio, China), and the nuclei were stained by 4′6-Diamidino-2-phenylindole-dihydrochloride (DAPI, Beyotime Institute of Biotechnology, Shanghai, China). The images were taken using a fluorescence microscope (Eclipse 80i, Nikon, Japan), and the vascular tube length and number were calculated using NIS-Elements AR 3.1 (Nikon).

PF127 powder was slowly added to the complete α-MEM (containing 20% FBS, 100 μg/ml of penicillin, and 100 μg/ml of streptomycin) to prepare a 20% (w/v) solution. The solution was placed on a shaking table at 4°C for 24 h to ensure that the powder was fully dissolved. Then, the PF127 solution was filtered by using a .22 μm pore size bottle-top filter and stored at 4°C for future use. Our previous study confirmed that the PF127 solution could preserve a solution state at 4°C and transform a hydrogel state at body temperature (Albashari et al., 2020).

The biocompatibility of PF127 hydrogel was evaluated using a Live/Dead Viability/Cytotoxicity Kit (Invitrogen, CA, United States) and Cell Counting Kit-8 (CCK-8, Dojindo Molecular Technologies) assay. In brief, the cultured DPSCs were trypsinized and resuspended in PF127 solution on ice. Then, 200 μl of cell-PF127 mixture solution containing 5 × 10⁴ cells/ml was added to a 48-well plate. After seeding, the plate was kept at 37°C in 5% CO₂ for 5 min to induce gel formation. The fresh medium was then added to each well, and the plate was transferred to a cell incubator. After being incubated for 1, 3, and 5 days, the Calcein-AM/pyridine iodide reagent mixture solution was added to the plates and incubated in the dark at 37°C for 10 min and observed using a fluorescence microscope (Eclipse 80i, Nikon, Japan). 10% CCK-8 solution was added to each well and transferred to a cell incubator for another 2 h to quantitatively analyze the cell proliferation activity. Absorbance at 450 nm was measured using an absorbance microplate reader (Varioskan LUX, Thermo Fisher Scientific).

All experiments were performed under sterile conditions. Female rabbits were anesthetized by a mixture of 2% isoflurane in 70% nitrous oxide and 30% oxygen and maintained with 1% isoflurane via a face mask. Then the skin was prepared, and a 4–5 cm incision was made to expose the FT and ovaries. The junction between the FT and the uterus was clamped with an arterial clip. .1 ml of absolute ethanol was injected into the FT using a 1 ml syringe. After 2 min, the arterial clip was loosened, and the FT was wiped around using a moist gauze and sutured. After 4 h, the HE results indicated that the FT had evident damage, and the structure of mucosa folds collapsed (Supplementary Figure S1), indicating that the animal model of FT injury was successfully constructed.

Twenty four female white rabbits were randomly assigned to four groups: 1) sham group without any treatment; 2) FT injury (FTI) group that directly injected saline into the FT after the effect of anhydrous alcohol for 4 h; 3) DPSCs group that received an intravenous injection of DPSCs at the ear margin after the effect of anhydrous alcohol for 4 h; 4) PF127-encapsulated DPSCs (DPSCs-PF127) group that injected the mixture of DPSCs and PF127 into the FT after the effect of anhydrous alcohol for 4 h. After treatment for 4 weeks, the rabbits were euthanized by intraperitoneal injection of 3% pentobarbital sodium and the samples were collected and used for future analysis.

The FT specimens were fixed in 4% formaldehyde, embedded in paraffin, sectioned into slides, stained with hematoxylin and eosin (H&E) and Masson’s trichrome, and recorded using an optical microscope (ECLIPSE 80i, Nikon, Tokyo, Japan).

The TUNEL apoptosis assay kit (Beyotime, Nantong, Jiangsu, China) assay was performed as previously described (Zhu et al., 2020). In brief, the sections were dewaxed in xylene and rehydrated with gradient ethanol. Then, the sections were treated with TUNEL-FITC (1:200), and the nuclei were stained with DAPI for 10 min. Images were taken using a fluorescence microscope (Eclipse 80i, Nikon, Japan).

Immunofluorescence was performed to assess the primary antibody of VEGF-A and KI67 according to the manufacturer’s procedures. The sections were treated with primary antibodies, namely, VEGF-A (1:200, Abcam, Cambridge, United Kingdom) and KI67 (1:200, Abcam, Cambridge, United Kingdom), overnight. Then, the sections were incubated with fluorogenic secondary antibodies for 1 h at 37°C. Finally, the sections were washed with PBS and incubated with DAPI for 10 min. For immunohistochemistry, the sections were treated with a primary antibody, namely, Vimentin (VIM; 1:200, Abcam, Cambridge, United Kingdom), overnight at 4°C and then incubated with horseradish-peroxidase-conjugated secondary antibody for another 2 h at 37°C. Then, the sections were developed using 3,3′-diaminobenzidine. All sections were observed using a fluorescence microscope (Eclipse 80i, Nikon, Japan).

The concentrations of TNF-α and IL-10 in serum were measured using TNF-α (CSB-E06998Rb, CUSABIO Biotech Co., Ltd., Wuhan, China) and IL-10 (CSB-E06897Rb, CUSABIO Biotech Co., Ltd., Wuhan, China) ELISA kits. The blood of rabbits was collected; the serum was obtained after centrifugation, and the serum level of inflammatory cytokines was detected in accordance with the manufacturer’s protocols. All of the samples were repeated.

After 4 weeks, the FT was collected. Total RNA was extracted from 50 mg of FT using a TRIzol reagent (Invitrogen, CA, United States). One microgram of RNA was reverse transcribed into cDNA using a PrimeScript™ RT reagent kit gDNA Eraser (RR047A, TaKaRa BIO Inc., Kusatsu, Shiga, Japan). Real-time PCR was performed using the SYBR Premix EX TaqTM II. cDNA was amplified using the primer sequences shown in Table 1. The PCR cycling parameters used were as follows: 95°C for 10 s, annealed at 95°C for 5 s and extended at 60°C for 30 s for 40 cycles, followed by thermal denaturation. The mRNA expression was evaluated on the basis of SYBR green fluorescence for 18 s as the endogenous control.

Graph pad 7.0 statistics was used to process the data, and the results were presented as means ± standard errors. Statistical differences were obtained by Student’s t-test and one-way analysis of variance, followed by Tukey’s test or Dunnett post hoc test. p < .05 was considered as significant differences.

The dental pulp is a soft connective tissue containing mesenchymal tissues, which is located in the medullary cavity of the tooth. DPSCs were extracted from the dental pulp, and the procedure for obtaining dental pulp tissues is shown in Supplementary Figure S2A. DPSCs had great cell proliferation capacity, and they can complete the whole primary culture for about 20 days (Supplementary Figure S2B). DPSCs possessed MSC-like properties and positively expressed MSC-like surface markers, including CD166, CD73, CD90, and CD44 (Supplementary Figures S3A, B), but they negatively expressed hematopoietic lineage markers, including CD19, CD34, and HLA-DR (Supplementary Figure S3A). Moreover, DPSCs had an excellent capacity for multilineage differentiation, and they could form mineralized nodules by positive staining of alizarin red S in the induced differentiation of osteoblasts. DPSCs were also induced to differentiate into adipogenesis with the formation of lipid droplets, which were stained with oil red O. In addition, DPSCs could differentiate into chondrocytes and show the evident staining of Alcian Blue (Supplementary Figure S3C).

The effects of DPSCs on the secretion of inflammatory factors in LPS-induced RAW264.7 cells were evaluated by immunofluorescence staining and ELISA. The results indicated that the expression of TNF-α and IL-10 was upregulated in macrophages by LPS stimulation for 24 h. The expression level of TNF-α was reduced in the presence of DPSCs (Figures 1A, B), whereas the expression of IL-10 was significantly increased (Figures 2A, B). As shown in Figures 1C, 2C, the results indicated that the level of TNF-α and IL-10 in supernatants was also increased after LPS induction for 24 h. In addition, DPSCs could downregulate the expression of TNF-α and upregulate the expression of IL-10, and they showed a significant difference compared with the LPS-induced one.

The angiogenic characteristics of DPSCs were evaluated by tubular network formation. The directly induced DPSCs (d-DPSCs), which were 3D cultured with GelMA hydrogel in the completed EGM-2™ medium, could form capillary-like networks. In addition, the number and diameter of closed networks were higher in d-DPSCs than in DPSCs cultured with α-MEM medium, and the difference was statistically significant (Figures 3A, B). Moreover, our results indicated that DPSCs could enhance the capillary-like structure formation of HUVECs using the indirectly cocultured system. Meanwhile, the number and length of capillary-like structures were higher in HUVECs cocultured with DPSCs than in HUVECs cultured alone (Figures 3C, D).

As shown in Figure 4A, PF127 hydrogel demonstrated a thermosensitive characteristic. It was clear and liquid solution when it was cold and transited to a homogeneous semi-solid hydrogel status at 37°C. Its gelation time was 80 ± 5 s. The biocompatibility of PF127 hydrogel containing DPSCs was evaluated using Live/Dead and CCK-8 assays. As for the Live/Dead assay, the results indicated that PF127 hydrogel had great cytocompatibility. After being cocultured with DPSCs, most of the cells were alive and stained with green from day 1 to day 5. In addition, the ratio of live cells had no difference between day 3 and day 5, but all of them were higher than that of day 1 (Figures 4B, C). The proliferation of encapsulated DPSCs in PF-127 hydrogel was detected for 5 days by CCK-8 assay. The results showed that DPSCs could continue to proliferate in PF127 hydrogel from day 1 to day 5, and the difference was statistically significant (Figure 4D).

Four weeks after surgery, upon macroscopic view, injured part of the FT tissue in the FTI and DPSCs groups shrunk greatly (red arrow). In contrast, injured FT tissue in the DPSCs-PF127 group appeared normal and was similar to that of in the sham group (Supplementary Figure S4). Meanwhile, the cross-section of the middle site of the FT tissue was stained with H&E and Masson’s trichrome among the experimental groups. As shown in Figure 5A, the H&E-staining results showed that the normal FT tissue had large mucosa-fold-filled lumen and loosely thick muscularis wall in the sham group. Moreover, the mucosal fold was composed of a single columnar epithelium and a branched core of vascular tissues. In the FTI group, the mucosal fold disappeared, and several infiltrated inflammatory cells and densely thin muscle layers could be observed in the damaged tube tissue. After treatment with DPSCs alone or DPSCs-PF127 in situ injection, the FT gradually returned to its normal structure, and mucosal folds with multi-branching recovered, but inflammatory infiltration also occurred in the lumen and muscle wall in the DPSCs group. In the DPSCs-PF127 group, the structure of mucosal folds and muscle walls was almost similar to that of the sham control group. As for Masson’s trichrome staining in each group (Figure 5B), the results showed thin and sparse collagen fibers, which were stained with blue dispersed in the submucosal and muscle wall in the sham group. However, massive collagen fiber infiltration occurred, which was stained with deep blue in the muscle wall and lumen in the FTI group. After treatment with DPSCs alone, a moderate amount of collagen fibers (blue) scattered in the muscle wall, some of which were infiltrated in the lumen. In the DPSCs-PF127 group, the FT preserved its normal structure, and the muscle wall and mucosal fold were filled with a small number of collagen fibers, which had been shown in blue.

The effect of DPSCs on FT cell apoptosis was analyzed by TUNEL staining. As shown in Figure 6A, TUNEL staining showed evident green fluorescence in the FTI group, indicating that FT cells underwent significant apoptosis. However, a small amount of TUNEL staining cells were found in the sham group. After treatment of DPSCs, TUNEL staining with green fluorescence gradually reduced, but the expression of TUNEL in the DPSCs group was stronger than that in the DPSCs-PF127 group. The number of TUNEL staining cells was highest in the FTI group and lowest in the DPSCs-PF127 group (Figure 6B). The number of TUNEL staining cells had no significant difference between the sham group and DPSCs-PF127 group (p > .05).

For immunofluorescence staining and immunohistochemistry analysis, the expression of a cell proliferation marker (KI67) and endothelial cell markers (VEGF-A and VIM) was investigated to evaluate the effect of DPSCs on FT cell proliferation and vascularization in vivo. The fluorescence expression of KI67 and VEGF-1 in the DPSCs and DPSCs-PF127 groups was stronger than that in the FTI group. All of the markers were highly expressed in the DPSCs-PF127 group and moderately expressed in the DPSCs and sham groups but weakly expressed in the FTI group (Figures 7A, 8A). In the three experimental groups, the intensity of KI67 and VEGF-A-positive regions was presented in the following order: DPSCs-PF127 > DPSCs > FTI. Meanwhile, a statistically significant difference was found between the DPSCs-PF127 group and the sham group (p < .01, Figures 7B, 8B). For immunohistochemistry (Figure 9), the results showed that VIM was highly expressed in the DPSCs-PF127 group and moderately expressed in the sham group but weakly expressed in the DPSCs and the FTI groups (Figure 9A). The intensity of VIM-positive regions was highest in the DPSCs-PF127 group compared with that of the other groups, but no statistically significant difference was found between the DPSCs group and FTI group (p > .05, Figure 9B).

ELISA was used to evaluate the serum levels of TNF-α and IL-10 in each group. The results showed that the serum TNF-α levels in the FTI and DPSCs groups were significantly higher than those in the sham and DPSCs-PF127 groups (p < .001). Meanwhile, no statistically significant difference was observed between the DPSCs-PF127 and sham groups (p > .05). However, the serum IL-10 levels in the FTI group were lower than those in the other groups (p < .01), and no statistically significant difference was observed among these groups except for the FTI group (p > .05, Figure 10A). For real-time PCR (RT-PCR) analysis (Figure 10B), the results indicated that OVGP, IL-10, and KI67 mRNA were highly expressed in the DPSCs-PF127 group and moderately expressed in the DPSCs and sham groups but weakly expressed in the FTI group. However, no significant difference was observed between the sham and DPSCs groups (p > .05). In addition, caspase 3 mRNA was weakly expressed in the DPSCs-PF127 group and highly expressed in the TFI group. Moreover, the expression level of caspase 3 showed a significant difference between the TFI group and other groups (p < .05).