Hypertrophy in mesenchymal stem cell chondrogenesis: effect of TGF-beta isoforms and chondrogenic conditioning

Abstract

Induction of chondrogenesis in mesenchymal stem cells (MSCs) with TGF-beta leads to a hypertrophic phenotype. The hypertrophic maturation of the chondrocytes is dependent on the timed removal of TGF-beta and sensitive to hypertrophy-promoting agents in vitro. In this study, we have investigated whether TGF-beta3, which has been shown to be more prochondrogenic compared to TGF-beta1, similarly enhances terminal differentiation in an in vitro hypertrophy model of chondrogenically differentiating MSCs. In addition, we tested the impact of the time of chondrogenic conditioning on the enhancement of hypertrophy. MSCs were chondrogenically differentiated in pellet culture in medium containing TGF-beta1 or TGF-beta3. After 2 or 4 weeks, chondrogenic medium was switched to hypertrophy-inducing medium for 2 weeks. Aggregates were analyzed histologically and biochemically on days 14, 28 and 42. The switch to hypertrophy medium after 14 days induced hypertrophic cell morphology and significant increase in alkaline phosphatase activity compared to the chondrogenesis only control using both TGF-beta1 and TGF-beta3. After 28 days predifferentiation, differences between hypertrophic and control groups diminished compared to 14 days predifferentiation. In conclusion, chondrogenic conditioning with both TGF-beta isoforms similarly induced hypertrophy in our experiment and allowed the enhancement of the hypertrophic chondrocyte phenotype by hypertrophic medium. Enhancement of hypertrophy was seen more clearly after the shorter chondrogenic conditioning. Therefore, to utilize this experimental model as a tool to study hypertrophy in MSC chondrogenesis, a predifferentiation period of 14 days is recommended.

Fig. 1

Cartilage matrix production in chondrogenic pellet cultures of MSCs monitored as sulfated glycosaminoglycan content normalized to DNA content. Three independent experiments were performed with cells isolated from 3 different donors (a, b donor 1; c, d donor 2; e, f donor 3) treated with TGF-β1 (a, c, e) or TGF-β3 (b, d, f). Open columns: chondrogenic control group (CM); grey columns: induction of hypertrophy after 14 days pre-differentiation (Hyp 14d); dark columns: induction of hypertrophy after 28 days pre-differentiation (Hyp 28d). Error bars show standard deviation; * significantly different from chondrogenic control (p < 0.05); n = 4. Reproducible significant differences between the two TGF-β isoforms were not detected. Significant differences between hypertrophic and control groups were only detected in one cell population with TGF-β3 (f).

Fig. 2

Hypertrophy in chondrogenic pellet cultures monitored on the basis of ALP activity in the culture medium. Three independent experiments were performed with cells isolated from 3 different donors (a, b donor 1; c, d donor 2; e, f donor 3) treated with TGF-β1 (a, c, e) or TGF-β3 (b, d, f). Open columns: chondrogenic control group (CM); grey columns: induction of hypertrophy after 14 days pre-differentiation (Hyp 14d); dark columns: induction of hypertrophy after 28 days pre-differentiation (Hyp 28d). Error bars show standard deviation; * significantly different from chondrogenic control (p < 0.05); n = 8. Significantly higher ALP activity was measured in the culture medium in the hypertrophic groups on day 21, day 28 and day 35 in all cell populations and on day 42 in one cell population (donor 2; c, d). No reproducible significant differences were detected between the two TGF-β isoforms.

Fig. 3

Histological and histochemical analysis of chondrogenic pellet cultures of MSCs treated with TGF-β1. DMMB (a–e), ALP (f–j) and von Kossa (k–o) staining of control chondrogenic cultures on day 14 (a, f, k), day 28 (b, g, l) and day 42 (d, i, n) and of day 28 cultures with hypertrophic treatment started on day 14 (c, h, m) and day 42 cultures with hypertrophic treatment started on day 28 (e, j, o). Selected day 28 and day 42 cultures for von Kossa staining (l, m) also received β-glycerophosphate supplementation starting on day 14 for day 28 cultures and on day 28 for day 42 cultures. DMMB staining shows the typical hypertrophic morphology with large lacunae in the hypertrophic group on day 28 (c). ALP activity is located in the periphery of cultures in standard chondrogenic medium (f, g, i) and positive throughout the entire aggregates in hypertrophic medium (h, j). Mineralization upon addition of β-glycerophosphate is restricted to the periphery of both control and hypertrophic cultures (l–o).

Fig. 4

Immunohistochemical analysis of matrix composition of chondrogenic pellet cultures of MSCs treated with TGF-β1. Staining for type I (a–e), type II (f–j) and type × collagen (k–o) of control chondrogenic cultures on day 14 (a, f, k), day 28 (b, g, l) and day 42 (d, i, n) and of hypertrophic cultures on day 28 after 14 days of chondrogenic pre-differentiation (c, h, m) and day 42 after 28 days of chondrogenic pre-differentiation (e, j, o). Strong type I collagen staining in the periphery of the cultures (a–e). Type I collagen staining in the center of the aggregates is stronger in hypertrophic (c, e) than control cultures (a, b, d). Initial type II collagen expression is detected on day 14 (f) and strong staining is observed on day 28 and day 42 in both control and hypertrophic cultures (g–j). Type × collagen staining is negative on day 14 (k) and positive in the hypertrophic (m, o) and in the control cultures (l, n) on day 28 and day 42.