(453b) 2D Culture and 3D Material Arrays to Define Optimal Conditions for Tenogenic MSC Differentiation

The repair of tendon injuries is critically limited by the inability to produce large numbers of tendon fibroblasts, or tenocytes. Multipotent cells, such as mesenchymal stem cells (MSCs), are a potential cell source for tendon repair and can be driven to differentiate down osteogenic, chondrogenic, adipogenic, and other lineages using established protocols. While previous work has highlighted the possible roles of mechanical stimulation, cell confinement/alignment, and co-culture with primary tenocytes, there is no standard protocol for robust tenogenic MSC differentiation. One reason for this is the lack of molecular markers to distinguish tendon from other mesenchymal tissues. Recently, scleraxis (SCX) and thrombospondin-4 (THBS4) have been suggested as tendon-specific markers. Here we present 2D and 3D culture systems to evaluate combinations of biochemical and biophysical factors to drive tenogenic MSC differentiation.

We first used a 2D culture system to assess the effects of a wide range of biochemical factors on the expression of tenogenic markers SCX and THBS4. Human MSCs were seeded in 96-well plates at a density of 1 x 10³ cells/well and cultured for 14 days in the presence of single or binary combinations of tenogenic soluble factors bFGF, IGF-1, GDF-5, and GDF-7. Additionally, the influence of standard differentiation agents dexamethasone and ascorbic acid was tested. MSCs were fixed, incubated overnight with primary antibodies against SCX and THBS4, tagged with fluorescently labeled secondary antibodies, and then assayed on a spectrophotometer. Fluorescence was normalized to cell number (DAPI). We found that the presence of dexamethasone and ascorbic acid resulted in increased levels of both SCX and THBS4 on a per cell basis for nearly every soluble factor combination assayed. Additionally, the IGF-1 and GDF-7 groups elicited the greatest increase in SCX and THBS4 production per cell.

In parallel, we have developed arrays of 3D collagen-GAG (CG) scaffolds. Tenocytes are known to de-differentiate in standard 2D culture so we used this system to interrogate the combined effect of scaffold microstructure and biochemistry on tenogenic MSC differentiation. Scaffold arrays with two distinct microstructural regions were fabricated by controlling local heat transfer during the freeze-drying process. A specialized mold was created for this purpose, consisting of a polysulfone chip (2 mm thick) with circular holes (6.5 mm diameter) mounted on a removable base. The geometry/spacing of the nodes was designed to be identical to the dimensions of a 96-well plate. The removable base contained an aluminum section and a polysulfone section (k_aluminum/k_polysulfone ~ 900); this disparity in thermal conductivity was intended to control local heat transfer, and subsequently ice crystal growth kinetics, during freezing. The mean pore sizes for scaffolds were 88 ± 28 μm within the aluminum section and 157 ± 47 μm within the polysulfone section of the array for a freezing temperature of -40ºC. Cells seeded in the scaffold array and stained with calcein showed no significant differences in fluorescence compared with cells seeded in a standard 96-well plate. Within the array we could also pattern near-linear (R² = 0.91) step-wise gradients of a model protein (biotinylated concanavalin A) using carbodiimide chemistry. Ongoing work with these systems is continuing to define optimal biochemical and biophysical cues for guiding tenogenic MSC differentiation.