Human adipose-derived stem cells (hADSCs) were purchased from Guangzhou Saliai Stem Cell Science and Technology Co. Ltd. and human SCs were purchased from The Cell Bank of the Type Culture Collection of The Chinese Academy of Sciences. The identification method of hADSCs used were as follow: i) cell survival rate ≥80%; ii) microorganism detection (bacteria: negative, Mycoplasma: negative, endotoxin <0.25 EU/ml); iii) special human-derived virus (HIV: negative, HBV: negative, HCV: negative, TP-DNA: negative, HCMV: negative, EBV: negative); iv) cell surface markers (CD73 ≥95%, CD90 ≥95%, CD105 ≥95%, CD45 ≤2%, CD34 ≤2%, HLA-DA ≤2%, CD11b ≤2%, CD19 ≤2%); v) differentiation abilities (Oil red O staining was positive after adipogenic differentiation for 21 days; Alizarin red S staining was positive after osteogenic differentiation for 21 days). hADSCs were maintained in low glucose-DMEM (Gibco; Thermo Fisher Scientific, Inc.) containing 10% FBS (Gibco; Thermo Fisher Scientific, Inc.), 2 mM glutamine (Merck KGaA), 1% non-essential amino acids (Gibco; Thermo Fisher Scientific, Inc.), 55 mM 2-mercaptoethanol (Bio-Rad Laboratories, Inc.), 1 mM sodium pyruvate (Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin (Beyotime Institute of Biotechnology) and 100 mg/ml streptomycin (Beyotime Institute of Biotechnology). SCs were cultured in SC medium (ScienCell Research Laboratories, Inc.) with 10% FBS. Cells were cultured at 37°C in a humidified incubator with 5% CO₂. hAMSCs were induced to differentiate into iSCs as previously described (36). Following exosome co-culture for 24 h, the cells were harvested and washed with PBS. In addition, the stem cell research underwent an EMRO process according to the guidelines of the International Society for Stem Cell Research (37).

A total of 24 adult male SD rats (weight, 200–250 g; age, 8-weeks) were anesthetized with pentobarbital (50 mg/kg) as previously described (38). The animals were purchased from Guizhou Laboratory Animal Engineering Technology Center. Animals were kept in an environmentally controlled breeding room (temperature: 20±2°C; humidity: 60±5%; 12 h dark/light cycle) with sterilized food and clean water. At 1 cm above the bifurcation into the tibial and common fibular nerves, the sciatic nerve was exposed and cut using ophthalmic scissors. The incision was closed with 4–0 nylon sutures. Subsequently, ~1 cm NT-3 chitosan conduit (Guangzhou Beogene Biotech Co., Ltd.) filled with iSC hydrogel (100 µl; 5×10⁷ cells) was used to promote sciatic nerve repair. The NT-3 chitosan conduit was sutured at both ends of the disconnected sciatic nerve to connect them. The in vivo experiment was divided into the following four experimental groups (n=6): i) normal (sham operation); ii) iSCs (sciatic nerve cut and filled with iSC hydrogel); iii) conduit (sciatic nerve cut and NT-3 chitosan conduit connection); and iv) iSCs + conduit (sciatic nerve cut and NT-3 chitosan conduit connection filled with iSC hydrogel). Sciatic function, step length and stride length (39) were measured at 4, 8 and 12 weeks of treatment. After anesthesia with pentobarbital (100 mg/kg) by intraperitoneal injection, cervical dislocation was used as the euthanasia method, and death was confirmed after complete cardiac arrest and pupil dilation. Human intervention was implemented to prevent or relieve unnecessary pain and disease if considered endpoints were met at any moment during the experiment: i) 20% body weight loss; ii) the occurrence of other complications. All experimental procedures involving animals were approved by the Institutional Animal Care and Use Committee of Zunyi Medical Hospital [(2017)2-035].

The pcDNA3.1-vector and pcDNA3.1-SOX2 were constructed and purchased from Guangzhou RiboBio Co., Ltd. Small interfering (si)RNAs were purchased from Angon Biotech Co., Ltd. The sequences of the siRNAs are as follows: si-SOX2-1, 5′-UGCAUCAUGCUGUAGCUGCCG-3′; si-SOX2-2, 5′-UGUCCAUGCGCUGGUUCACGC-3′; si-FN1-1, 5′-ACAAACUGGAGGUUAGUGGGA-3′; si-FN1-2, 5′-UUUAUCUGAUAGUGUUUUCCA-3′ and negative control (NC), 5′-UUCUCCGAACGUGUCACGUTT-3′. The cells were transfected with siRNAs or overexpression plasmids using Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Transfected cells were cultured at 37°C with 5% CO₂ for 48 h.

RT-qPCR was performed as previously described (40). In brief, total RNA was extracted using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Total RNA (1 µg) was reverse transcribed into cDNA using PrimeScript™ RT Reagent kit (Takara Bio, Inc.). Subsequently, qPCR was performed in triplicate for each sample using TB Green^® Fast qPCR Mix (Takara Bio, Inc.) with GAPDH as the internal reference gene. The following reaction conditions were applied: 3 min at 95°C, 40 cycles of 15 sec at 95°C and 30 sec at 58°C and maintenance at 4°C. The following primers were used: SOX2 forward, 5′-GCCGAGTGGAAACTTTTGTCG-3′ and reverse, 5′-GGCAGCGTGTACTTATCCTTCT-3′; FN1 forward, 5′-CGGTGGCTGTCAGTCAAAG-3′ and reverse, 5′-AAACCTCGGCTTCCTCCATAA-3′; GAPDH forward, 5′-GGAGCGAGATCCCTCCAAAAT-3′ and reverse, 5′-GGCTGTTGTCATACTTCTCATGG-3′.

Western blotting was performed as previously described (40). In brief, after washing with PBS, cells and tissues were lysed in RIPA lysis buffer (Beyotime Institute of Biotechnology). Protein quantification was performed using a BCA assay (Beyotime Institute of Biotechnology). Proteins were mixed with 5X Loading buffer (Beyotime Institute of Biotechnology). The following primary antibodies were used: NF160 (1:1,000; cat. no. ab254348; Abcam), GFAP (1:5,000; cat. no. ab7260; Abcam), CD31 (1:1,000; cat. no. ab9498; Abcam), SOX2 (1:1,500; cat. no. ab92494; Abcam), FN1 (1:1,000; cat. no. 15613-1-AP; ProteinTech Group, Inc.) and GAPDH (1:15,000; cat. no. 60004-1-Ig; ProteinTech Group, Inc.). The membranes were incubated with anti-rabbit IgG HRP-conjugated (1:2,000; cat. no. 7074S; Cell Signaling Technology, Inc.) and anti-mouse IgG HRP-conjugated (1:2,000; cat. no. 7076S; Cell Signaling Technology, Inc.) secondary antibodies for 2 h. Protein bands were then visualized using an ECL chemiluminescent kit (Thermo Fisher Scientific, Inc.) and semi-quantified using ImageJ v1.8 software (National Institutes of Health). The numbers under the band of western blotting were the semi-quantitative results measured using ImageJ v1.8 software.

Cells were seeded into 6-well plates and cultured to the sub-confluent state. After starvation in serum-free SC medium for 24 h, a straight wound was scratched at the bottom of the plate with a 200-µl sterile pipette tip. After gently rinsing with PBS, cells were cultured in serum-free medium for 24 or 48 h. Cell migration was observed and calculated at 0, 24 and 48 h using an inverted light microscope. Scratch-healing was calculated as follows: Scratch healing (%)=(initial scratch area-final scratch area)/initial scratch area ×100.

Transwell assays were performed using Transwell chambers (Corning, Inc.). Cells were seeded (7.5×10³) into the upper compartment in SC medium with 1% FBS, whereas the lower chamber was filled with SC medium with 10% FBS. Invading cells were fixed and calculated after 24 h.

Cell viability was assessed using the CCK-8 assay. Cells were seeded (5×10³ cells/well) into a 96-well plate. After 24 h, the cells were treated. After 48 h, the cells were incubated with 100 µl SC medium containing 10% CCK-8 reagent (Dojindo Molecular Technologies, Inc.) for 1.5 h at 37°C. Then absorbance was measured at a wavelength of 450 nm using a microplate reader (Tecan Group, Ltd.). Cell viability was calculated as follows: Cell viability (%)=[optical density (OD) treatment-OD blank)/(OD control-OD blank) ×100.

Cell proliferation was determined using an EdU assay. Cells were seeded into a 24-well cell culture plate and then 10 µM EdU was added to each well. After incubation for 2 h at 37°C, the cells were fixed with 4% paraformaldehyde for 30 min at room temperature. After washing with PBS, the incorporated EdU was detected using a EdU kit (Beijing Solarbio Science & Technology Co., Ltd.) for 30 min at room temperature. Subsequently, the cells were stained with DAPI for 5 min at room temperature and then visualized using a fluorescence microscope (Olympus Corporation).

SCs with SOX2 overexpression RNA-seq data (GSE94590) were obtained from the Gene Expression Omnibus database (41). Then, the values in the Sox2 overexpression group was compared with the normal group to obtain the differentially expressed genes, respectively. The analysis was performed using limma package (http://bioinf.wehi.edu.au/limma/) in R. The threshold for significant was set at 1.2 or 0.83 fold change and P<0.05. A Venn diagram (http://bioinformatics.psb.ugent.be/webtools/Venn/) was used to identify the common differential genes in Sox2 clone1 and Sox2 clone2 group. Then, the common differential genes were used for Kyoto Encyclopedia of Genes and Genome (KEGG) analysis in the David database (https://david.ncifcrf.gov/).

iSCs were cultured in iSC induction medium containing 10% exosome-depleted FBS. Exosomes were extracted using Ribo Exosome Isolation Reagent (for cell culture media) (Guangzhou RiboBio Co., Ltd.) according to the manufacturer's protocol. A Nanosight LM10 HS-BF (Nanosight Ltd.) was used to measure the size and concentration of the purified exosomes (42). CD81 and CD63 (43), as exosome markers, were used to identified the exosomes using flow cytometry. The following monoclonal antibodies were used: CD63 (1:50; cat. no. GTX41877; GeneTex, Inc.) and CD81 (1:50; cat. no. GTX34568; GeneTex, Inc.). Exosome pellets were resuspended in PBS and frozen at −80°C. The exosome pellet was used for subsequent experiments. The exosomes were added to the medium with exosome-free FBS and then co-cultured with SCs for 24 h.

Data are presented as the mean ± SD of three independent experiments. Statistical analyses were performed using SPSS software (version 19.0; IBM Corp.). All cell experiments were independently repeated at least in triplicate. Comparisons between two groups were analyzed using an unpaired Student's t-test. One-way ANOVA was adopted to compare multiple groups. Post hoc multiple comparisons were made using Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The rat sciatic nerve injury model was used to simulate clinical peripheral nerve injury. hAMSCs were induced into iSCs in vitro. At 1 cm above the bifurcation into the tibial and common fibular nerves, the sciatic nerve was exposed and cut using ophthalmic scissors. An NT-3 chitosan conduit filled with iSC hydrogel was used to promote sciatic nerve repair (Fig. 1A). The in vivo experiment was divided into the following four groups: i) normal (sham operation); ii) iSCs (sciatic nerve cut and filled with iSC hydrogel); iii) conduit (sciatic nerve cut and NT-3 chitosan conduit connection); iv) iSCs+conduit (sciatic nerve cut and NT-3 chitosan conduit connection filled with iSC hydrogel). The maximum percentage of body weight loss was 16% throughout the whole experiment. To investigate the repair effect of the sciatic nerve, the gastrocnemius muscle weight was measured. As shown in Fig. 1B, the weight of the gastrocnemius muscle in the iSCs+conduit group was significantly heavier compared with that in the conduit or iSCs groups. The sciatic function index, step length and stride length data demonstrated recovery in nerve function. At 4, 8 and 12 weeks, measurements were taken. The results indicated that iSCs with the NT-3 conduit promoted sciatic nerve recovery. Compared with the iSCs or conduit groups, the iSCs+conduit group effectively promoted the repair of the sciatic nerve (Fig. 1C). The results indicated that iSCs with the NT-3 chitosan conduit promoted sciatic nerve recovery.

SCs with SOX2 overexpression RNA-seq data (GSE94590) were obtained from the Gene Expression Omnibus database. Differential genes identified by Venn analysis are presented in Fig. 2A. There were 501 different genes in both SOX2 clone 1 and SOX2 clone 2 SCs. The differential genes were then analyzed using Kyoto Encyclopedia of Genes and Genomes enrichment (KEGG) analysis. The results identified that SOX2 was related to ‘focal adhesion process’, ‘PI3K-AKT signaling pathway’ and ‘ECM-receptor interaction’ (Fig. 2B). It is critical for cell proliferation and migration. In addition, cell proliferation and migration are the cytological basis of fibronectin fibrillogenesis. SOX2 and FN1 protein expression in gastrocnemius muscle tissues was measured. The results indicated that SOX2 and FN1 levels were decreased in the iSCs or conduit groups compared with those in the normal group. However, the expression levels were increased in the iSCs+conduit group compared with those in the iSCs or conduit groups (Fig. 2C). In addition, SOX2 and FN1 protein expression levels were more abundant in iSCs compared with those in hAMSCs (Fig. 2D). These results suggest that SOX2 participates in SC proliferation and migration.

SOX2 was overexpressed and knocked down using the SOX2 overexpression plasmid (pcDNA3.1-SOX2) and si-SOX2, respectively. SOX2 mRNA expression was increased following SOX2 overexpression plasmid transfection, but decreased after si-SOX2 transfection (Fig. 3A). hAMSCs were induced to differentiate into iSCs as previously described. The transfection efficiency was confirmed for subsequent experiments. Scratch wound assays were conducted to evaluate cell migration after SOX2 transfection. The distance between iSCs cells induced by stem cells is larger than that of ordinary cells, consistent with a study by Feng et al (44). The migration rate was significantly increased after SOX2 overexpression, but significantly decreased after SOX2 knockdown in the iSCs (Fig. 3B). In addition, the migratory ability was analyzed using Transwell assays. After 24 h of seeding, the number of migratory SOX2-overexpression or -knockdown iSCs in the lower chamber was counted. The number of cells in the lower chamber was significantly increased with SOX2 overexpression, but significantly decreased with si-SOX2 transfection (Fig. 3C). The aforementioned results indicated that SOX2 promoted the migratory ability of iSCs. In addition to cell migration, cell proliferation was assessed to evaluate nerve repair. The CCK-8 assay was performed to evaluate the effect of SOX2 on cell viability in iSCs. The results showed that SOX2 overexpression significantly increased iSC viability (Fig. 3D). The protein expression levels of neurofilament 160 (NF-160), glial fibrillary acidic protein (GFAP) and CD31 were measured to evaluate nerve cell activity. The protein expression levels of NF-160, GFAP and CD31 were increased with SOX2 overexpression. However, the opposite trend was observed with si-SOX2 transfection (Fig. 3F). Cell proliferation was examined by performing EdU assays. EdU labelling is indicated by green fluorescence and the cell nucleus is shown in blue. The results indicated that SOX2 promoted iSC proliferation (Fig. 3E). Taken together, these results indicated that SOX2 increased the migration and proliferation of iSCs.

SOX2 was overexpressed and knocked down using the SOX2 overexpression plasmid and si-SOX2 in the iSCs, respectively. FN1 mRNA and protein expression levels were increased following SOX2 overexpression, but decreased following SOX2 knockdown (Fig. 4A and B). Therefore, the results indicated that FN1 expression may be regulated by SOX2. FN1 was knocked down in SOX2-overexpressing iSCs. The protein expression levels of NF-160, GFAP and CD31 were detected using western blotting. The protein expression levels of NF-160, GFAP and CD31 were decreased after FN1 knockdown in SOX2-overexpression iSCs. The result indicated that FN1 inhibited SOX2 overexpression-mediated improvements in nerve cell viability (Fig. 4C). In addition, cell migration was measured by performing Transwell assays. The number of migratory cells was significantly increased with SOX2 overexpression, whereas the number was significantly decreased with si-SOX2 transfection. Compared with the SOX2 overexpression group, the number of migratory cells was decreased in the SOX2-overexpression iSCs with FN1 interference (Fig. 4D). Furthermore, scratch wound assays were performed to detected the migratory ability of SOX2-overexpression iSCs with FN1 interference. The results were consistent with the aforementioned results, in which FN1 interference inhibited SOX2 overexpression-mediated improvements in the migration ability of iSCs (Fig. 4E). Cell proliferation was measured by performing EdU assays under the same treatment. As presented in Fig. 4F, cell proliferation was elevated by SOX2 overexpression. However, FN1 interference decreased SOX2 overexpression-mediated improvements in iSC proliferation. Taken together, these results indicated that SOX2 upregulated FN1 expression to promote iSC migration and viability.

Exosomes were isolated from iSCs using ultracentrifugation. Next, the effect of the exosomes isolated from iSCs on SC proliferation and migration was investigated. From Genecards database (https://www.genecards.org/), we know that CD81 and CD63 are located not only in the plasma membrane, but also in other organelles, such as the endosome, cytosol, and lysosome. Western blot analysis can be used to detect CD81/CD63 in all cells. However, flow cytometry can detect CD81/CD63 in the cell membrane to identify the type of cell. The surface markers (CD63 and CD81) of iSC exosomes were assessed using flow cytometry to confirm the successful isolation of exosomes from iSCs. The percentages of CD63 and CD81 were increased from 50% to 83.5 and 86.8%, respectively (Fig. 5A). The protein expression levels of NF-160, GFAP and CD31 were elevated in the SCs with exosome treatment or iSC co-culture (Fig. 5B). The exosome treatment group and iSC co-culture group displayed decreased wound gap widths (Fig. 5C). Taken together, the results indicated that the effect of exosomes isolated from iSCs was similar to that of iSC co-culture on cell migration. SCs cells were treated with si-SOX2 or si-NC alone with exosomes isolated from iSCs, and migration ability was measured by wound scratch assay. The functional rescue results indicated that SOX2 and exosomes isolated from iSCs promoted migration in the SCs. In some sense, SOX2 promotes SC migration via exosomes (Fig. 5D). In addition, the CCK-8 assay was performed to assess the effect of exosomes isolated from iSCs on SC viability. Exosomes secreted by iSCs significantly elevated the viability of SCs (Fig. 5E). Furthermore, cell proliferation was enhanced in SCs treated with exosomes (Fig. 5F). These results indicated that exosomes secreted by iSCs promote SC viability and migration.