SOX9 gene transfer via safe, stable, replication-defective recombinant adeno-associated virus vectors as a novel, powerful tool to enhance the chondrogenic potential of human mesenchymal stem cells

Abstract

Introduction: Transplantation of genetically modified human bone marrow-derived mesenchymal stem cells (hMSCs) with an accurate potential for chondrogenic differentiation may be a powerful means to enhance the healing of articular cartilage lesions in patients. Here, we evaluated the benefits of delivering SOX9 (a key regulator of chondrocyte differentiation and cartilage formation) via safe, maintained, replication-defective recombinant adeno-associated virus (rAAV) vector on the capability of hMSCs to commit to an adequate chondrocyte phenotype compared with other mesenchymal lineages.

Methods: The rAAV-FLAG-hSOX9 vector was provided to both undifferentiated and lineage-induced MSCs freshly isolated from patients to determine the effects of the candidate construct on the viability, biosynthetic activities, and ability of the cells to enter chondrogenic, osteogenic, and adipogenic differentiation programs compared with control treatments (rAAV-lacZ or absence of vector administration).

Results: Marked, prolonged expression of the transcription factor was noted in undifferentiated and chondrogenically differentiated cells transduced with rAAV-FLAG-hSOX9, leading to increased synthesis of major extracellular matrix components compared with control treatments, but without effect on proliferative activities. Chondrogenic differentiation (SOX9, type II collagen, proteoglycan expression) was successfully achieved in all types of cells but strongly enhanced when the SOX9 vector was provided. Remarkably, rAAV-FLAG-hSOX9 delivery reduced the levels of markers of hypertrophy, terminal and osteogenic/adipogenic differentiation in hMSCs (type I and type X collagen, alkaline phosphatase (ALP), matrix metalloproteinase 13 (MMP13), and osteopontin (OP) with diminished expression of the osteoblast-related transcription factor runt-related transcription factor 2 (RUNX2); lipoprotein lipase (LPL), peroxisome proliferator-activated receptor gamma 2 (PPARG2)), as well as their ability to undergo proper osteo-/adipogenic differentiation. These effects were accompanied with decreased levels of β-catenin (a mediator of the Wnt signaling pathway for osteoblast lineage differentiation) and enhanced parathyroid hormone-related protein (PTHrP) expression (an inhibitor of hypertrophic maturation, calcification, and bone formation) via SOX9 treatment.

Figure 1

Detection of rAAV-mediated transgene expression in undifferentiated monolayer and chondrogenically differentiated aggregate cultures of human mesenchymal stem cells (hMSCs). Cells were transduced with rAAV-lacZ or rAAV-FLAG-hSOX9 (a) in monolayer culture (40 μl each vector) or (b) in aggregate cultures (100 μl each vector), as described in Materials and methods, or left untreated, and processed to monitor transgene expression 21 days after vector application by analyzing the immunoreactivity to the FLAG tag or to SOX9. (a) Anti-FLAG at magnification ×20 and anti-SOX9 at magnification ×4; (b) magnification ×4.

Figure 2

Histologic analyses in chondrogenically differentiated aggregate cultures of human mesenchymal stem cells (hMSCs). Aggregate cultures were prepared and transduced with rAAV-lacZ or rAAV-FLAG-hSOX9, as described in Figure 1, or left untreated, and processed on day 21 for histologic staining with H&E, toluidine blue, and alizarin red, as described in Materials and methods. All at magnification ×4.

Figure 3

Expression analyses in differentiated aggregate cultures of human mesenchymal stem cells (hMSCs). Aggregate cultures were prepared and transduced with rAAV-lacZ or rAAV-FLAG-hSOX9, as described in Figure 1, or left untreated, and processed on day 21 for (a) immunodetection of type II, type I, and type X collagen (all at magnification ×4), and (b) gene-expression analysis by real-time RT-PCR amplification after total cellular RNA extraction and cDNA synthesis, as described in Materials and methods. The genes analyzed included the transcription factor SOX9, and types II, I, and X collagen (COL2A1, COL1A1, COL10A1), alkaline phosphatase (ALP), matrix metalloproteinase 13 (MMP13), osteopontin (OP), the transcription factor RUNX2, β-catenin, parathyroid hormone-related protein (PTHrP), lipoprotein lipase (LPL), and the peroxisome proliferator-activated receptor gamma 2 (PPARG2), with GAPDH serving as a housekeeping gene and internal control (primers are listed in Materials and methods). Ct values were obtained for each target and GAPDH as a control for normalization, and fold inductions (relative to untreated aggregates) were measured by using the 2^-ΔΔCt method. Statistically significant compared with (a) condition without vector treatment or (b) rAAV-lacZ.

Figure 4

Analyses in osteogenically and adipogenically differentiated cultures of human mesenchymal stem cells (hMSCs). Cells in monolayer cultures were transduced with rAAV-lacZ or rAAV-FLAG-hSOX9 (100 μl each vector) or left untreated and induced toward osteogenic or adipogenic differentiation, as described in Materials and methods. Cultures were processed on day 21 for (a) ALP staining (osteogenesis; magnification ×4) and Oil Red O staining (adipogenesis; magnification ×10) and (b) and (c) gene-expression analysis with real-time RT-PCR amplification, as described in Figure 3. The genes analyzed included ALP, COL1A1, OP, and RUNX2 for osteogenically differentiated cultures (b) and LPL and PPARG2 for adipogenically differentiated cultures (c), with GAPDH serving as a housekeeping gene and internal control in both cases. Ct values were obtained for each target and GAPDH as a control for normalization, and fold inductions (relative to untreated cultures) were measured by using the 2^-ΔΔCt method. Statistically significant compared with (a) condition without vector treatment or (b) rAAV-lacZ.