All experimental procedures involving animals in the present study conformed with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals, and were approved by the Administration Committee of Experimental Animals (Jiangsu, China). A total of 40 male adult New Zealand white rabbits (weight, 2.5–3.0 kg; age, ~3 months) were purchased from the Jiangsu Academy of Agricultural Sciences (Jiangsu, China) for use in the present study. The rabbits were maintained in cages at the Animal Center of Southeast University (Nanjing, China) at 24°C under a 12-h light/dark cycle with ad libitum access to food and water. A total of 30 rabbits were used to establish IDD models, and were subsequently randomly divided into three groups: 200 μl PRP or phosphate-buffered saline (PBS; Sigma-Aldrich, St. Louis, MO, USA) were injected into the degenerated discs (PRP group and PBS group, respectively) using a 26-gauge needle; 200 μl PRP-containing BMSCs was also injected (PRP-BMSC group). Another 10 rabbits without intervention were used as the control group. The present study was approved by the ethics committee of Southeast University.

The BMSCs were isolated from a New Zealand white rabbit, as described previously (17). Briefly, 2 ml iliac bone marrow was collected from the rabbits and immediately mixed with 2 ml PBS. Following size-fractionation using a Ficoll-Paque Plus (1.077 g/ml; GE Healthcare Life Sciences, Shanghai, China), BMSCs were isolated by centrifugation at 400 × g for 30 min at room temperature. The isolated BMSCs were seeded (10⁵ cells/cm²) into a 25 cm² culture flask (Corning, Inc., Acton, MA, USA), and cultured in Dulbecco's modified Eagle's medium with low glucose (DMEM-LG; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), supplemented with 10% (v/v) fetal bovine serum (FBS; Wisent., Inc., St-Jean-Baptiste, QC, Canada) at 37°C under a 5% CO₂ atmosphere. The culture medium was replaced every 2–3 days and BMSCs after three passages were used in the present study. The expanded BMSCs were induced towards adipogenic, osteogenic and chondrogenic lineages, as previously described (18).

BMSCs (2×10³ cells/cm²) at passage three were cultured in 6-well culture plates containing the adipogenic induction medium, which consisted of 10% (v/v) FBS, 200 μM indomethacin, 0.5 mM 3-isobutyl-1-methylxanthine (IBMX) 1 μM dexamethasone, 10 μg/ml insulin (all: Cyagen Biosciences, Inc., Santa Clara, CA, USA) and DMEM-LG. The culture medium was changed every 3–4 days. After 4 weeks of induction, the formation of oil droplets was tested using Oil Red O staining, in which the induced BMSCs were fixed in 70% ethanol for 10 min and stained with 0.3% fresh Oil Red O solution (Sigma-Aldrich) for 2 h prior to observation under a microscope (Leica DMRXA2; Leica Microsystems GmbH, Wetzlar, Germany).

BMSCs (2×10³ cells/cm²) at passage three were cultured in 6-well culture plates containing the osteogenic induction medium, which consisted of 10% (v/v) FBS, 100 nM dexamethasone, 10 mM β-glycerophosphate (Sigma-Aldrich), 0.05 mM ascorbic acid (Sigma-Aldrich) and DMEM-LG. The culture medium was changed every 3–4 days. After 4 weeks of induction, the formation of calcium nudes were examined by Alizarin red staining, in which the induced BMSCs were fixed with 70% ethanol for 10 min and stained with 0.5% alizarin red solution (pH 4.1; Sigma-Aldrich) for 2 h, prior to observation under a microscope (Leica DMRXA2).

A pellet culture system was applied to confirm the chondrogenesis of the isolated BMSCs. Briefly, 2.5×10⁵ cells were centrifuged at 450 × g for 10 min at room temperature to form a small pellet and cultured in 15 ml polypropylene tubes (Sigma-Aldrich). The chondrogenic induction medium, including 10 nM dexamethasone, 10 ng/ml TGF-β3 (Cyagen Biosciences, Inc.), 50 mg/ml ascorbic acid and 50 mg/ml ITS+Premix (Cyagen Biosciences, Inc.) in DMEM with high glucose was used, and replaced every 3 days. Following culturing in the chondrogenic induction medium at 37°C under a 5% CO₂ atmosphere for 4 weeks, the pellets were fixed in formalin (Sigma-Aldrich) and sectioned for staining with 1% toluidine blue solution (Seebio Biotech, Inc., Shanghai, China) for 2 h. Following washing with PBS, the sections were mounted using neutral gum (China Sinopharm International, Co., Ltd., Shanghai, China) for observation under a microscope (Leica DMRXA2).

The PRP was prepared, according to a method described by Landesberg et al (19) with modifications. Briefly, under general anesthesia with pentobarbitone sodium (30 mg/kg; Sigma-Aldrich), 6 ml fresh blood was obtained from the central artery of the rabbits ear using a syringe containing 0.6 ml acid citrate dextrose-A solution (Kermel Chemical Reagent Co., Ltd., Tianjin, China) as a anticoagulant. Subsequently, 0.2 ml whole blood was obtained to perform a platelet count. The obtained whole blood was primarily centrifuged by a centrifuge (SC-04; Zhongke Meiling Cryogenics Co., Ltd., Hefei, China) at 200 g for 10 min. Subsequently, the plasma fraction was collected and further centrifuged at 1,000 g for another 10 min. The upper three-quarters of supernatant plasma was carefully removed, and the remaining plasma and platelets were gently agitated and designated as PRP (~0.6 ml). Subsequently, 0.2 ml of the prepared PRP was drawn for platelet counting, and the remaining PRP (~0.4 ml) was retained for further use. To prepare the PRP-BMSC mixture, the BMSCs were trypsinized (Sigma-Aldrich) and diluted in PBS to 1×10⁶ cells/ml, and 0.2 ml volumes of cell suspension were carefully mixed with the previously prepared PRP (~0.4 ml).

The rabbit lumbar IDD models of the PRP, PBS and PRP-BMSC groups were established by needle puncture into the annulus fibrosis (Fig. 1). Briefly, under general anesthesia, the L4–5 and L5–6 lumbar discs were exposed using a contralateral side approach under a sterile environment. Following puncture of the discs with a 21-gauge needle to the depth of 5 mm, the surgical wound was immediately sutured. To confirm early degeneration of the punctured discs, the spines of the rabbits were evaluated using a 3.0-T MRI scan 2 weeks following post-needle puncture), the targeted discs were exposed again for the respective treatments. At weeks 1, 2, and 8 following the different treatments, the targeted discs of the groups were scanned using 3.0-T MRI. At week 8 following intervention, the rabbits in each group were sacrificed by intravenous injection with an overdose of pentobarbitone sodium (120 mg/kg). The vertebral body-disc-vertebral units of the targeted segments were carefully cut off and fixed in 10% (v/v) formalin in neutral buffer for paraffin (Sigma-Aldrich) embedding. The paraffin blocks were sectioned at a thickness of 5 μm. Type II collagen immunohistochemistry and hematoxylin and eosin (H&E; Sigma-Aldrich) staining were performed to evaluate the histological changes in the treated discs. At 200× magnification, five visual fields were randomly selected from each sample for evaluation under a microscope (Leica DMRXA2). The Image-Pro Plus 6.0 image analysis system (Media Cybernetics, Inc., Rockville, MD, USA) was utilized to analyze the integrated absorbance (IA) value of the yellow-stained type II collagen.

MRI was performed using a 3.0T imager unit (Verio 3.0T MR; Siemens Medical Solutions, Erlangen, Germany). Following general anesthesia with pentobarbitone sodium (30 mg/kg), the rabbits were placed in the supine position on a quadrature surface coil for MRI scanning. The parameters of the T2-weighted images were as follows: Time-to-repeat=1,800 ms; time-to-echo=71 ms; field of view=140×140 mm; slice thickness=2 mm; averages=9; image matrix=288×384. The signal intensity of the discs was evaluated using T2-weighted images. IDD grade was evaluated using the modified Thompson classification (20) of grades I–IV, as follows. Grade I, normal signal intensity; grade II, minimal decreased signal intensity without obvious narrowing of high signal area; grade III, moderate decrease of signal intensity; grade IV, severe decrease of signal intensity. Two senior radiologists evaluated the images to reach a consensus.

Statistical analysis was performed using SPSS 20.0 software (IBM SPSS, Armonk, NY, USA). Data of all treatment groups are presented as the mean ± standard error of the mean and were analyzed using one way analysis of variance. Pairwise comparisons were analyzed using a Student-Newman-Kuells test. The level data of the MRI images were analyzed using a Mann-Whitney U-test. P<0.05 was considered to indicate a statistically significant difference.

At 3 weeks post-induction, the isolated BMSCs (Fig. 2A) were successfully induced toward adipogenic, osteogenic and chondrogenic lineages. The osteogenic differentiation potential of the isolated BMSCs was confirmed using Alizarin-red staining (Fig. 2B). Lipid droplets were formed and confirmed using Oil Red O staining (Fig. 2C). As for chondrogenic induction of the BMSCs, the cell pellets sectioned for toluidine blue staining (Fig. 2D) were positive, indicating the chondrogenic potential of the isolated BMSCs.

The platelet concentration of the prepared PRP was 776±48×10⁹/l, whereas the concentration in the whole blood was only 159±21×10⁹/l. The PRP prepared in the present study contained almost five times the number of platelets in the whole blood.

Representative sagittal MRI images showed the variation of the targeted discs with high T2-weighted signal intensity in all groups (Fig. 3). T2 signal intensity was stable at each time point in the control group. At 2 weeks following needle puncture (week 0), the injected discs experienced mild signal loss. The treated discs of the PBS group experienced continuous T2 signal loss during the 8-week period, whereas the discs in the PRP-BMSC group showed an increase in signal intensity over time. PRP exhibited a regenerative effect on the degenerated discs at 1 and 2 weeks following treatment by increasing the signal intensity. However, at week 8 post-treatment, the discs of the PRP group exhibited decreased signal intensity. To determine the degenerative changes of the targeted discs, the mean MRI score of each group was presented (Fig. 4). At weeks 1 and 2, no significant differences were found in the MRI grading between the PRP and PRP-BMSC groups (P<0.05). At week 8, the PRP-BMSC group showed a significantly higher MRI grading score, compared with the PRP group (P<0.05).

In the control group, discs without any interventions exhibited a high density of ECM. At 8 weeks following the different treatments, the discs in the PRP (Fig. 5A) and PBS (Fig. 5B) groups exhibited decreased ECM. In the PRP-BMSC group (Fig. 5C), ECM and cell density were well preserved, however, fewer NP cells were observed in the PRP and PBS groups, compared with the control group (Fig. 5D).

Representative immunohistochemistry of type II collagen staining exhibited brown staining. The type II collagen staining of the targeted discs in the PRP group (Fig. 6A) became weakly positive, whereas the staining in the PRP-BMSC group (Fig. 6B) was strongly positive at week 8. As with the PRP group, the staining in the PBS group (Fig. 6C) became weakly positive at week 8, whereas staining in the control group was strongly positive (Fig. 6D). The IA values of the control group (100910.06±4319.57) and PRP-BMSC group (80620.30±5030.62) were significantly higher, compared with those of the PRP group (45682.05±3501.89) and PBS group (43768.22±2275.67), as shown in Fig. 7 (P<0.05).