Different tenogenic differentiation capacities of different mesenchymal stem cells in the presence of BMP-12

Abstract

Background: Mesenchymal stem cells (MSCs) are regarded as a promising cell-based therapeutic tool for tendon repair. This study aimed to compare the different tenogenic differentiation capacities of the three types of MSCs in the presence of bone morphogenic protein 12 (BMP-12).

Methods: MSCs were isolated from rat bone marrow (BM), inguinal adipose tissue (AD), and synovium (SM) from the knee joint. MSCs were characterized by morphology, proliferation, trilineage differentiation, and surface marker analysis. Tenogenic differentiation potential was initially assessed using real-time polymerase chain reaction, Western blot, and enzyme-linked immunosorbent assay in vitro. Histological assessments were also performed after subcutaneous implantation of BMP-12 recombinant adenovirus-infected MSCs in nude mice in vivo.

Results: The three types of MSCs exhibited similar fibroblast-like morphology and surface markers but different differentiation potentials toward adipogenic, osteogenic, and chondrogenic lineage fates. Bone marrow-derived MSCs (BM-MSCs) showed the most superior in vitro tenogenic differentiation capacity, followed by synovial membrane-derived MSCs (SM-MSCs) and then adipose-derived MSCs (AD-MSCs). After implantation, all three types of MSC masses infected with BMP-12 recombinant adenovirus emerged in the form of fiber-like matrix, especially in 6-week specimens, compared with the control MSCs in vivo. BM-MSCs and SM-MSCs revealed more intense staining for collagen type I (Col I) compared with AD-MSCs. Differences were not observed between BM-MSCs and SM-MSCs. However, SM-MSCs demonstrated higher proliferation capacity than BM-MSCs.

Conclusion: BM-MSCs exhibited the most superior tenogenic differentiation capacity, followed by SM-MSCs. By contrast, AD-MSCs demonstrated the inferior capacity among the three types of MSCs in the presence of BMP-12 both in vivo and in vitro.

Figure 1

Morphology and proliferation differences among three types of MSCs. a–c BM-MSCs, AD-MSCs, and SM-MSCs were isolated from the bone marrow, adipose tissue, and synovium, respectively. Images show the MSCs at passage 3. Magnification ×10, bar 50 µm. d All three types of MSCs were seeded in a 96-well plate at 5 × 10³ per well and were cultured for 1, 2, 3, 4, 5, and 6 days. The cell numbers at each day were measured using CCK-8 assay. The cells on day 0 were used as control. The data were obtained from three independent experiments, each performed in triplicate. Each bar represents mean ± SD (*p < 0.05, vs. BM-MSCs, at the same time point, ANOVA).

Figure 2

Comparative analysis of the tri-lineage differentiation capacities of the three types of MSCs. a All three types of MSCs were cultured in adipogenic differentiation medium for 14 days and were assessed by oil red O staining (used to identify adipogenic differentiation). Magnification ×10, bar 50 µm. b and c All three types of MSCs were cultured in osteogenic differentiation medium for 21 days and were assessed by ALP (alkaline phosphatase, a byproduct of osteoblast activity) staining (b) and Alizarin red staining (used to identify calcium deposits) (c). Magnification ×4, bar 50 µm. d All three types of MSCs were cultured in chondrogenic differentiation medium for 21 days and were assessed by immunohistochemical staining with anti-type II collagen antibody. Magnification ×4, bar 50 µm. e PPAR-γ (adipogenic differentiation marker) was evaluated by real-time RT–PCR at 0, 3, and 7 days post induction (mean ± SD, n = 3, *p < 0.05, ANOVA). f and g OCN (osteogenic differentiation marker) (f) and Col II (chondrogenic differentiation marker) (g) were evaluated by real-time RT–PCR at 0, 7, and 14 days post induction (mean ± SD, n = 3, *p < 0.05, ANOVA). h–j Analysis of oil red O staining (h), ALP staining (i), and alizarin red staining (j). The ratio of positive-stained area to the total area of cells was calculated using Image-Pro Plus 6.0 software (mean ± SD, n = 5, *p < 0.05, ANOVA). k Analysis of immunohistochemical staining of collagen type II. Expression levels were quantified by mean intensity of the images analyzed using Image-Pro Plus 6.0 software (mean ± SD, n = 5, *p < 0.05, ANOVA).

Figure 3

Immunophenotypic profiles of MSCs derived from three different rat tissues. Three types of MSCs at passage 3 (1 × 10⁶) were incubated with FITC-conjugated CD105, CD73, CD90, and CD45 antibodies (Abcam, Cambridge, UK) in 1% FBS/PBS for 1 h. Cells were washed three times with 1% FBS/PBS and then resuspended in 500 μL of PBS. For negative controls, FITC-conjugated nonspecific IgG fractions (Abcam) were substituted for the primary antibodies.

Figure 4

Influence of recombinant Ad-BMP-12 virus on cell viability. a All three types of MSCs were infected with Ad-BMP-12 at MOIs of 50, 100, 200, and 500. Untreated MSCs were used as controls. Flow cytometry was used to assess infection efficiency. The data were obtained from three independent experiments, each performed in triplicate. Each bar represents mean ± SD (*p < 0.05, vs. control, t test). b All three types of MSCs were seeded in a 24-well plate at 1 × 10⁵ per well and infected with Ad-BMP-12 at MOI of 500. Cell numbers were measured using CCK-8 assay on days 1 and 3. The data were obtained from three independent experiments, each performed in triplicate. Each bar represents mean ± SD (*p < 0.05, vs. control, t test). c Images were acquired using a fluorescence microscope to show infection efficiency and cell viability on days 1 and 3. Cells with green fluorescence were the infected MSCs. Magnification ×10, bar 50 µm.

Figure 5

Comparative analysis of the tenogenic differentiation capacities of three types of MSCs. a–c MSCs were cultured with Ad-BMP-12 for 1, 3, and 7 days. The expression levels of the specific tenogenic genes Scx (a), Tnmd (b), and Tnc (c) were evaluated by real-time RT–PCR at 0, 1, 3, and 7 days after infection (each bar represents mean ± SD from three independent experiments; *p < 0.05, ANOVA). The BM-MSCs infected with Ad-GFP on day 0 were used as control. d Western blot was performed with total proteins from all three types of MSCs at 0, 1, 3, and 7 days after infection to detect the levels of Scx, Tnmd, and Tnc. β-Tublin was used as control. e Col I expression was evaluated by real-time RT–PCR at 0, 1, 3, and 7 days after induction (each bar represents mean ± SD from three independent experiments; *p < 0.05). The BM-MSCs infected with Ad-GFP on day 0 were used as control. f Col I concentration in the cell lysate was quantified at 0, 1, 3, and 7 days post Ad-BMP-12 infection by using a Col I ELISA kit (each bar represents mean ± SD from three independent experiments; *p < 0.05, ANOVA).

Figure 6

Properties of the implanted Ad-BMP-12 infected cell masses. a MSCs were initially treated with Ad-BMP-12 for 1 day and were then subcutaneously injected into the nude mice for 3 or 6 weeks. The MSCs treated with Ad-GFP were used as control. Images show the morphologies of the cell masses after 3 and 6 weeks from injection. b and c Comparison of the three types of MSCs in terms of the area (b) and weight (c) of the cell masses. (Each bar represents mean ± SD; *P < 0.05, statistical method: t test: 3w/6w BMP-12 group vs. 3w/6w control group, respectively; ANOVA: 3w/6w BMP-12 group in BM-MSC vs. 3w/6w BMP-12 group in AD-MSC vs. 3w/6w BMP-12 group in SM-MSC, respectively).

Figure 7

H&E staining of the implanted cell masses. The specimens were fixed and embedded in paraffin. Sections of 5 µm cut through the center of the cell masses were stained with hematoxylin/eosin (H&E). Images show the staining at 3 weeks (a) and 6 weeks (b). Magnification ×10, bar 200 µm.

Figure 8

Immunohistochemical analysis of implanted MSCs. a and b The specimens were fixed and embedded in paraffin. Sections of 5 µm cut through the center of the cell masses were stained with anti-Col I antibody. Images show the staining at 3 weeks (a) and 6 weeks (b). Magnification ×10, bar 200 µm. c Immunohistochemical staining of Col I of rat Achilles tendon was used as positive control. d Col I expression was quantified by mean intensity of the images analyzed using Image-Pro Plus 6.0 software. Achilles tendon was used as positive control (each bar represents mean ± SD, n = 5; *p < 0.05, ANOVA).