Fibrinogen (F3879, Sigma, United States) dissolved in phosphate-buffered saline (PBS, 0.01 M, pH 7.40) at a concentration of 200 mg/ml, thrombin (T4393, Sigma, United States) dissolved in PBS at a concentration of 100 U/ml, and genipin (G4796, Sigma, United States) dissolved in dimethyl sulfoxide (D2650, Sigma, United States) at a concentration of 400 mg/ml, were prepared for the glue fabrication. This component concentration of Fib-T-G gel was optimized by testing the biocompatibility and mechanical properties with different concentrations of genipin. The Fib-T-G gel contained final concentrations of fibrinogen 140 mg/ml, thrombin 28 U/ml, and genipin 6 mg/ml–these concentrations exhibited the best consistency for biocompatibility and adhesive strength for the following experiment in our pilot study, which was added as Supplementary Data. hMSCs (Passage 2, 15901; purchased from ScienCell Research Laboratories, Carlsbad, CA) routine cultured in Mesenchymal Stem Cell medium (7501, Sciencell, United States) which included DMEM supplemented with 5% fetal bovine serum (FBS, 0025, Sciencell, United States), 1% penicillin-streptomycin (PS, 0503, Sciencell, United States) and Mesenchymal Stem Cell Growth Supplement (MSCGS) (7552, Sciencell, United States) up to passage 6, were used in all the experiments in this study.

All animal experiments were performed according to the Guidelines for Care and Use of Laboratory Animals established by the Affiliated Hospital of Integrated Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine. The stroke-bearing protocol was approved by the Committee on the Ethics of Animal Experiments of the Hospital (Ethical Lot Number: AEWC-20191010-82). The experimental procedures were approved by the hospital ethic committee. Twenty male computer-randomized SD rats (average weight 450 ± 20 g) were utilized in this study with 10% chloral hydrate anesthesia. Four caudal intervertebral discs of each rat were exposed from the ventral side for four experimental groups respectively. In brief, a 2 mm width and 2.5 mm depth lesion was punctured by a custom-made blade. Fib-T-G MSCs group: annulus lesion filled and connected by 7 μL Fib-T-G glue mixed with 1.2×10³ hMSCs. Fibrinogen MSCs(Fib MSCs) group: annulus lesion filled by 7 μL fibrinogen mixed with 1.2×10³ hMSCs and connected carefully by one micro suture (ETHICON 7-0). Un-repair control: annulus lesion without repair. Intact control: discs exposed without any intervention. All the rats were fed with standardized food and moved freely, and were all sacrificed at 5 weeks.

After the operation, the specimens were fixed with 4% neutralized formalin (BL-G001, Nanjing Senbeijia Biotechnology Co., Ltd, China), then decalcified in 10% nitric acid (USP364, CPA, France). Each segment were separated and transferred to 75% ethanol (64-17-5, Nanjing Chemical Reagent Co., Ltd, China), then embedded in paraffin, cut in the mid-sagittal plane to 5-μm thickness, and stained with hematoxylin and eosin (H&E) for cellular constituents. Histological score of disc degeneration and annulus fibrosus were evaluated based on the following four categories of degenerative changes according to previous study (Masuda et al., 2005): annulus fibrosus, border between the anulus fibrosus and nucleus pulposus, cellularity of the nucleus pulposus, and matrix of the nucleus pulposus. IVD scores range from a normal disc as 4 points to a severely degenerated disc as 12 points, and AF scores range from a normal annulus being 1 point to a more than 30% ruptured or serpentined patterned fibers having 3 points. Picrosirius red (ab150681, Abcam, United States) was stained and assessed under a polarized light microscope for lamellar structure and continuity of collagen. Diaminobenzidine (DAB, ZLI-9018, ZSGB-BIO, China) staining of the main components of annulus fibrosus collagen 1 (COL1) (1:200, ab34710, Abcam, United States), collagen 2 (COL2) (1:200, ab34712, Abcam, United States), mechanical stimulation related protein RhoA (1:100, 10749-1-AP, Proteintech, China) and ROCK1 (1:200, ab97592, Abcam, United States) were conducted according to the product instructions and viewed under a microscope (IX73, Olympus, Japan).

The hMSCs were seeded into elastin precoated six-well BioFlex plates (Flexcell International Corporation) with 2.0×10⁶ cells/well and cultured in growth media (Dulbecco’s modified Eagle medium (DMEM, Gibco, United States) with 10% fetal bovine serum (FBS, Invitrogen, United States) and 1% penicillin-streptomycin (PS, Gibco, United States)) to reach confluence. Then cells were cultured by chondrocyte differentiation medium according to a previous study (Masuda et al., 2005; Lysdahl et al., 2013) (DMEM medium; 10% FBS; 100 nM Dexamethasone, D1756, Sigma; 50 μg/ml Ascorbic Acid-2phosphate, A4544, Sigma; 40 μg/ml L-Proline, P0380, Sigma; 1:100 ITS Supplement, 10201, Cyagen Biosciences Company; 10 ng/ml TGF-β1, 96-100-21-10, PeproTech; 1% PS), and the fresh medium was changed every 3 days. Cells were differentiated in the following two different mechanical conditions. The mechanical load group: cell cultures were loaded on the Flexercell FX 5000T cell-straining device with 3% strain and 1 HZ mechanical load stimulation for 1 h each day. The stable control group was cell cultures in a static environment.

Three cultures of each group were obtained after 7 days of differentiation. All the mediums in each well were collected to quantify the secreted sGAG and Collagens by Dimethylmethylene blue (DMMB, 341088, Sigma, United States) and Sicol assays (s1000, Biocolor, UK). Total RNA from samples was isolated using Trizol reagent (15596026, Invitrogen, United States) according to the manufacturer’s instructions. The RNA was synthesized by reverse transcription kit (E6560, NEB, United States) to cDNA, followed by quantitative real-time PCR (qPCR, ABI StepOnePlus, United States) analysis using SYBR Premix (E3003, NEB, United States) and ABI STEPONE PLUS. Gene expression of COL1A1, COL2A1, aggrecan, Mohawk, RhoA, ROCK1, and SOX9 were analyzed by qPCR. 3-phospate dehydrogenase (GAPDH) was used as a reference gene. The sequences of the genes are shown in Table 1.

Another three cultures of each group were obtained after 2 weeks of differentiation. Cell cultures were soaked in 4% paraformaldehyde in PBS with pH 7.4 for 10 min at room temperature and washed with ice-cold PBS three times. Samples were incubated with PBS containing 0.5% Triton X-100 for 10 min and subsequently washed in PBS three times. Then the samples added with 3% BSA in PBS were cultured for 30 min to block the unspecific binding of the antibodies. Samples incubated in the primary antibody COL1, COL2, Aggrecan (1:200, ab36861, Abcam, United States) in PBS-1% BSA solution in a humidified chamber overnight at 4 °C. After removing the primary antibody incubation solution, the cells were washed 3 times with PBS and then coincubated with DAB staining solution at room temperature. Picro Sirius red was stained to assess the alignment of collagen. Finally, all the samples were observed by a fluorescence microscope (IX73, Olympus, Japan).

The RhoA gene was transfected to hMSCs. In brief, hMSCs of 2.0×10⁶ cells/well were plated into six-well BioFlex culture plates precoated with elastin in culture media and allowed to reach 80% confluence. Transient transfections with the plasmids encoding wild-type RhoA (Public Protein/Plasmid Library, China) and the constitutively active Q63L RhoA were carried out in the RhoA using Lipofectamine™ LTX (15338100, Invitrogen, United States). Transfected cells were allowed to recover for 12 h, then trypsinized (25200, Gibco, United States), and sorted GFP-positive cells to purify only the cells containing transgenes. Sorted cells were centrifuged (1000 rpm/min), washed (PBS as washing solution), and prepared for further study.

RhoA transfected MSCs of 2.0×10⁶ each well were plated into six-well BioFlex culture plates and cultured in a differentiation medium (corresponding to 2.3), with normal MSCs culture as control. After 7 days, three samples of each group were obtained, and gene expression of COL1, COL2, Aggrecan, Mohawk, RhoA, and R was analysed by qPCR. After 2 weeks, three samples of each group were harvested, and the matrix of COL1, COL2, and aggrecan was determined by DAB staining as above study (corresponding to 2.3).

The hMSCs of 2.0×10⁶ each well were plated into six-well BioFlex culture plates and cultured in a differentiation medium. In mechanical load with inhibitor C3 group, 3% strain and 1 HZ mechanical load stimulation was conducted for 1 h each day, and Cell Permeable RhoA Inhibitor C3 transferase based (CT04, Cytoskeleton, United States) was added. In the mechanical load group, cells differentiated with a mechanical load but without inhibitor C3. In the control, cells differentiated without mechanical load and without inhibitor C3. After 7 days, we investigated the effect of mechanical strain on AF relative gene expression and the RhoA/ROCK1 by qPCR analysis.

The statistical significance for normally distributed data. Significance was indicated by letters in graphs. For gene expression data, a non-parametric distribution was assumed and presented as mean ± SD for an N ≥ 3. The data were analyzed by one-way ANOVA followed by multiple comparison tests. All tests were performed with GraphPad Prism 6.0 (GraphPad Software, La Jolla, CA, United States).

All the discs with lesions have degenerative signs with the fibrotic nucleus, annulus fissure, irregular fibrous tissue in HE and Sirius red staining, while the intact control has healthy loose nucleus matrix and regular annulus fibrous lamellae (Figures 2A–L). After histological disc degenerative scoring, the value 9.167 of the Fib-T-G MSCs group is significantly lower than the unrepaired control group 11.30 (p ≤ 0.01, and also lower than the 10.73 of the Fib MSCs group (p < 0.05) (Figure 2M). After histological annulus fissure degenerative scoring, the value 2.083 of the Fib-T-G MSCs group is significantly lower than the unrepaired control group 2.90 (p ≤ 0.01, and also lower than the 2.667 of the Fib MSCs group (p < 0.05). The lesion is still seen in both repaired groups but less in the Fib-T-G MSCs group at the middle sagittal histological section. The lesion area in the Fib-T-G MSCs group is apparently smaller than that in the Fib MSCs group (Figure 2N). Further, DAB staining in the repaired area of both repaired groups shows relatively more staining of collagen 1/2 (Figure 3) and RhoA/ROCK1 (Figure 4) in the Fib-T-G MSCs group, compared to the Fib MSCs group. Comprehensive results show that the Fib-T-G MSCs group with mechanical stimulation better affects repairing AF lesions.

As shown in Figure 5A, after 7 days of culture, the results showed the relative fibrocartilage differentiation gene expression of COL2, aggrecan, Mohawk, RhoA, ROCK1, and SOX9 in the mechanically stimulated environment were significantly higher than those in the static environment. After 2 weeks, the secreted sGAG and collagen contents were measured by Dimethylmethylene blue and Sicol assays (Figure 5B). The results showed that the secretion amount was also significantly higher in the mechanically stimulated environment than in the static environment. In addition, mechanical stimulation promotes hMSCs differentiation with more COL1, COL2, and Aggrecan secretion observed in immunohistostaining compared to the cell culture in the static environment (Figures 5C–H). At the same time, the alignment of collagen was observed by Sirius red staining (Figures 5I, J). Arranged collagen is seen in the mechanical stimulation group under polarized light observation but not in the static group. The comprehensive results demonstrate that the differentiation of MSCs towards AF tissue mechanical stimulation benefits the differentiation of hMSCs towards AF tissue, along with the RhoA gene activated.

As shown in Figure 6A, in the static environment, gene expression of COL1, COL2, Aggrecan, Mohawk, RhoA, ROCK1, and SOX9 in RhoA transfected hMSCs was significantly higher than that in normal non-RhoA transfected hMSCs after 7 days of culture. Relative more COL1, COL2, and aggrecan secreted by hMSCs by RhoA transfected hMSCs were observed by DAB staining (Figures 6B–G) after 2 weeks in culture. These results suggest that RhoA upregulation can promote AF-related gene expression and protein secretion. As shown in Figure 6H, in the mechanical stimulation environment, C3 Transferase inhibited RhoA signal hinders MSCs differentiation with significantly lower gene expression of collagen 2, aggrecan, Mohawk, ROCK1, SOX9 compared to the cell culture with no C3 Transferase after 7 days of culture. To sum up, without mechanical stimulation, overexpressing RhoA could promote differentiation; inhibiting RhoA reduced the differentiation effect with mechanical stimulation.