In Vitro Comparison of 2D-Cell Culture and 3D-Cell Sheets of Scleraxis-Programmed Bone Marrow Derived Mesenchymal Stem Cells to Primary Tendon Stem/Progenitor Cells for Tendon Repair

The poor and slow healing capacity of tendons requires novel strategies to speed up the tendon repair process. Hence, new and promising developments in tendon tissue engineering have become increasingly relevant. Previously, we have established a tendon progenitor cell line via ectopic expression of the tendon-related basic helix-loop-helix (bHLH) transcription factor Scleraxis (Scx) in human bone marrow mesenchymal stem cells (hMSC-Scx). The aim of this study was to directly compare the characteristics of hMSC-Scx cells to that of primary human tendon stem/progenitors cells (hTSPCs) via assessment of self-renewal and multipotency, gene marker expression profiling, in vitro wound healing assay and three-dimensional cell sheet formation. As expected, hTSPCs were more naive than hMSC-Scx cells because of higher clonogenicity, trilineage differentiation potential, and expression of stem cell markers, as well as higher mRNA levels of several gene factors associated with early tendon development. Interestingly, with regards to wound healing, both cell types demonstrate a comparable speed of scratch closure, as well as migratory velocity and distance in various migration experiments. In the three-dimensional cell sheet model, hMSC-Scx cells and hTSPCs form compact tendinous sheets as histological staining, and transmission electron microscopy shows spindle-shaped cells and collagen type I fibrils with similar average diameter size and distribution. Taken together, hTSPCs exceed hMSC-Scx cells in several characteristics, namely clonogenicity, multipotentiality, gene expression profile and rates of tendon-like sheet formation, whilst in three-dimensional cell sheets, both cell types have comparable in vitro healing potential and collagenous composition of their three-dimensional cell sheets, making both cell types a suitable cell source for tendon tissue engineering and healing.

of 420 nm with reference filter at 620 nm. CyQUANT cell proliferation kit was used as recommended by the manufacturer (Invitrogen, Eugene, OR, USA). HMSC-Scx and hTSPC (1.5×10 2 cells/cm 2 ) were plated in triplicates in 6-well dishes. Cells were frozen 0, 4, 8, 12 and 16 days of cell culture and stored at -80°C. Lastly, cells were lysed with the CyQUANT dye buffer for 5 mins and optical density was measured using the microtiter-plate reader (Thermo Scientific) at 480 nm excitation and 520 nm emission maxima. The WST-1 and CyQUANT experiments were reproduced thrice independently.

Self-renewal analysis
Colony-forming unit (CFU) assay was carried out as follows: 10 cells/cm 2 were plated in 10 cm cell culture dishes. After 14 days, formed colonies were stained with 0.5% crystal violet/methanol staining solution for 10 min and rinsed. Single colonies with 1-8mm diameter were counted and in the case of small cell clusters, a cluster of > 25 cells was considered a colony. Finally, CFU efficiency was determined with the formula CFU [%] = (number of colonies/number of plated cells) x100. CFU assays were performed with hMSC-Scx and hTSPC-in 2 different passages, and each experiment contained triplicate dishes.

Cell differentiation
For adipogenic differentiation, hMSC-Scx and hTSPC were plated in duplicates in 6-well dishes (5 x 10 3 cells/cm 2 ). Stimulation was started when cells reached full confluency. Cells were stimulated for 21 days using the induction medium composed of DMEM-high glucose medium, 10 % FBS, 1 μM dexamethasone, 0.2 mM indomethacin, 0.1 mg/ml insulin and 1 mM IBMX (all Sigma-Aldrich). Cells were grown for 5 days in induction medium, thereafter for 2 days in maintenance medium and then switched to induction medium again. The extent of adipogenic differentiation was evaluated by AdipoRed assay, which was implemented according to the manufacturer's instructions (Lonza, USA).
For chondrogenic differentiation, 2.5 x 10 5 hTSPC were seeded in V-bottomed 96-well polypropylene dishes and pellets were formed by centrifugation. Chondrogenic medium composed of DMEM-high glucose medium, 10 μM dexamethason, 1 mM sodiumpyruvat, 0.2 mM L-ascorbic-acid-2-phosphate, 1x ITS (all Sigma-Aldrich), 10 ng/ml TGF-β1 (R&D Systems) and 2 ng/ml BMP2 (R&D Systems) was applied onto the pellets for a culture period of 35 days. Pellets were fixed in 4 % paraformaldehyde and embedded in paraffin. Histological sections (7 μm thick) were stained with toluidine blue and collagen type II or aggrecan anitbodies using standard protocols. Collagen IIpositive area was measured with Image J program and expressed in % to pellet area.

Semiquantitative PCR
Total RNA was extracted from hMSC-Scx and hTSPC with RNeasy Mini Kit (Qiagen, Hilden, Germany), and 1 µg RNA was used for cDNA synthesis with AMV kit (Invitrogen) according to the manufacturer' instructions. Semi-quantitative PCR was performed with Taq DNA Polymerase