The neuronal differentiation of T-SCs was investigated in a three dimensional (3D) hybrid scaffold system by Patel et al[18]. This scaffold was fabricated by increasing the temperature of an aqueous solution of poly (ethylene glycol)-poly(L-alanine) to 37 °C, thus instigating the heat-induced sol-to-gel transition, in which T-SCs and growth factor-releasing microspheres were suspended. The gel exhibited a modulus of 800 Pa at 37 °C, similarly to that of brain tissue, and was robust enough to hold the microspheres and cells within the 3D cell culture. Neuronal growth factors were released over 12–18 d, and the encapsulated T-SCs gradually exhibited morphological changes from spherical to multipolar elongation. Significantly higher expression levels of neuronal biomarkers such as nuclear receptor-related protein, neuron-specific enolase, microtubule-associated protein-2, neurofilament-M, and glial fibrillary acidic protein were observed at both the mRNA and protein level in the hybrid system. This study clearly demonstrates the advantages of 3D hybrid scaffolds and highlights the importance of the sustained release of growth factors from hybrid systems to support the neuronal differentiation of T-SCs.

Schwann cells are the glial cells of peripheral nerves that wrap around the axons to form myelin in the peripheral nervous system. Schwann cells promote nerve regeneration by secreting trophic support molecules and establishing a supportive growth matrix[19]. Jung et al[20] demonstrated that T-SCs could be differentiated into Schwann-like cells over several steps. Briefly, T-SCs were induced to form neurospheres under stimulation with EGF, bFGF, and B27 for 7 d. These neurospheres were then triturated and re-plated onto laminin-coated dishes with Schwann cell differentiation medium. After 10 d of culturing, the cells exhibited morphological changes, including the formation of elongated bipolar and tripolar spindle shapes. Schwann-like cells differentiated from T-SCs highly express the Schwann cell markers GFAP, NGFR, S100B, KROX20, and KROX24. Notably, Schwann cells differentiated from T-SCs were able to produce myelinate axons in vitro when co-cultured with mouse dorsal root ganglion neurons. In a mouse model with a sciatic nerve injury, a marked improvement in gait and increased nerve regeneration were observed with Schwann-cell treatment. Therefore, T-SCs can be a useful source for Schwann cell-based cell therapy to treat neuropathic diseases.

For osteogenic differentiation, Choi et al[21] cultured T-SCs in commercially available osteogenic media (α MEM supplemented with 10% FBS, 0.1 mmol/L dexamethasone, 10 μmol/L β glycerophosphate, and 50 μg/mL ascorbic acid) for three weeks. Alizarin Red S staining confirmed the successful deposition of extracellular calcium in culture. During osteogenic differentiation, mRNA expression of osteocalcin decreased 0.28-fold after cryopreservation, whereas ALP expression showed no difference. This profile remained stable even after passage 15 (P15). Interestingly, the osteogenic differentiation of T-SCs increased with the number of passages, with the peak osteogenic potential observed for passage 10 (P10), which exhibited a 1.4-fold increase over P3[6]. The expression of CCN1, a gene that is closely related to the osteogenic differentiation of MSCs, increased at P10. This finding is consistent with previous studies that have reported that CCN1 expression modulates the osteogenic potential of MSCs by regulating the Wnt3A pathway[6,22].

Various scaffolds have been employed to enhance skeletal regeneration to replace damaged bone. Because bone tissue is highly vascularized, integration with the host tissue followed by subsequent angiogenesis is critical for successful treatment. As an example, Park et al[23] encapsulated T-SCs within highly water-swollen hydrogel through the sol-gel transition of the thermoresponsive polymer poly(ethylene glycol)-poly(L-alanine-co-L-phenyl alanine) (PEG-PAF). The encapsulated T-SCs were cultured in vitro in the presence of an osteogenic-induction medium. The osteogenic differentiation of T-SCs was investigated by evaluating the expression of osteogenic genes, including Runx 2, ALP, and OCN. With support from the hydrogel, osteogenic gene expression was two times higher than that of conventional tissue cultures without soluble factor supplements.

Incorporating specific functional groups into scaffold substrates is also known to affect the osteogenic differentiation of MSCs[24,25]. T-SCs were encapsulated in poly (ethylene glycol)-poly(L-alanine) diblock copolymer (PEG-L-PA) thermogel that was modified with the phosphate functional groups of polystyrene microspheres to facilitate the osteogenic differentiation of encapsulated MSCs. The osteogenic differentiation of tonsil-MSCs (T-MSCs) was analyzed, and all osteogenic biomarker expressions were significantly higher for the modified thermogel than for the native thermogel. Immunofluorescence staining also confirmed high OCN expression, and Kye et al[24] also found that, compared with thermogel modified with carboxylate group microspheres, the phosphate functional group (-PO43-) more readily induced the osteogenic differentiation of T-SCs.

The osteogenic differentiation of T-SCs can be further improved by the over-expression of BMP-2 through genetic material transfer. In Jeong et al[26], BMP-2 minicircle DNA vectors were employed to form nano-sized polyplexes with the CBA-106 polymer. CBA-106 is a bioreducible cationic poly (amido amine) that facilitates the intracellular delivery of genetic material. The osteogenic differentiation of T-SCs was investigated by evaluating the gene expression of osteogenic markers such as osteocalcin, Runx2, and Col 1. In vitro calcium deposition was also confirmed using Alizarin Red S staining on day 7 and day 14 of the cell culture. In addition, in vivo bone regeneration was attempted using T-SCs transfected with the BMP-2 gene. The T-SCs were 3D cultured using PLLA/PLGA scaffolds for one week and then transplanted into the skulls of immunodeficient mice. It was shown that bone regeneration increased 1.96-fold compared to the control group five weeks after treatment.

To mimic a cartilage-like microenvironment, T-SCs were encapsulated in thermogel consisting of the PEG-PAF block copolymer by Park et al[23]. Under specific medium conditions (25 μL chondrogenic supplement/2.5 mL basal medium), the much higher expression of Col II and sulfated glycosaminoglycan in T-SCs was achieved using the thermogel than using a monolayer culture. Moreover, unique branching was observed among the encapsulated cells within the hydrogel. These changes in cellular morphology may influence the chondrogenic differentiation of T-SCs during 3D culturing. In vivo studies also confirmed that T-SCs successfully undergo chondrogenic differentiation with high expression levels of biomarkers such as Col II, AGG, and Col X.

Kye et al[24] also investigated the chondrogenic differentiation of T-SCs within PEG-L-PA thermogel. To enhance the cellular attachment in the 3D hydrogel micro-environment, various polystyrene microspheres with thiol (-SH), phosphate (-PO3¬), carboxylate (-COO) and amino (-NH2) functional groups were incorporated into the thermogel. Of the incorporated microparticles, the PS-S, PS-P, and PS-C microspheres exhibited significantly higher COL II expression. COL II mRNA expression was much higher in the PS-S thermogel than in the PS-N thermogel even though the PS-S and the PS-N microspheres had similar sizes, clearly indicating that surface functional groups play an important role in stem cell differentiation.

Similarly, graphene oxide (GO) and reduced graphene oxide (rGO) were incorporated into PEG-L-PA thermogel by Park et al[27] to enhance surface functionality and thus provide good cellular adhesion. GO, or rGO (1 wt%) was suspended in a PEG- L-PA solution. T-SCs in GO/PEG-L-PA or rGO/PEG-L-PA were cultured in DMEM, and the spherical cellular morphology of the T-SCs was observed. When chondrogenic culture media enriched with TGF-β3 was utilized, T-SCs in hybrid systems aggregated extensively, and the expression levels of chondrogenic biomarkers such as SOX 9, COL II A1, COL II, and COL X increased. In particularly, COL II mRNA expression was 13 times higher in the GO/PEG-L-PA hybrid system than in the PEG-L-PA 3D system. Immunofluorescence analysis also revealed a significant increase in COL II expression and cell aggregation in the GO/PEG-L-PA hybrid system. In particular, the GO/PEG-L-PA 2D/3D hybrid system demonstrated the most significant increase in chondrogenic biomarker expression. These results indicate that the cooperative interaction among TGF-β3, COL II, and GO may be closely related to signaling cascades for chondrogenic differentiation.

Ryu et al[5] investigated T-SCs as a source for adipogenic differentiation due to their multi-lineage differentiation potential and self-renewal capacity by culturing them using commercially available adipogenic media (Gibco StemPro™ Adipogenesis Differentiation Kit, Thermo Fisher Scientific, Waltham, MA, USA) for three weeks. The adipogenic potential of the T-SCs was evaluated for P3, P7, P10, and P15. Interestingly, unlike other types of T-SC differentiation, the adipogenesis of T-SCs decreased continuously with passage number, with P10 approximately two-thirds that of P3.

Adipogenic differentiation of T-SCs in thermogel was also investigated using PEG-L-PA (molecular weight of each block: 1000−1080 Da) by Kye et al[24]. The differentiation potential of T-SCs was investigated by incorporating polystyrene microspheres with different functional groups into the hydrogel. mRNA expression and immune histochemical assays indicated that T-SCs preferentially underwent adipogenesis in ammonium (−NH3+)- or thiol (−SH)-functionalized thermogels, whereas chondrogenesis occurred predominantly in phosphate (PO32−)- or carboxylate (−COO−)-functionalized thermogels. This study thus suggests that the surface functional groups of microspheres can control the preferential differentiation of stem cells into specific cell types in 3D cultures.

In addition to functionalized hydrogels, Patel et al[28] reported that a composite system of GO and polypeptide thermogel (GO/P), prepared using the temperature-sensitive sol-to-gel transition of GO-suspended poly(ethylene glycol)-poly(L-alanine) (PEG-PA), significantly enhanced the expression of adipogenic biomarkers, including PPAR-γ, CEBP-α, LPL, AP2, ELOVL3, and HSL when compared with a native hydrogel system. It appears that insulin, an adipogenic differentiation factor, can preferentially adhere to the surface of the GO incorporated into thermogels; insulin is then slowly released into the environment in a sustained manner during the cell culturing period. In contrast, more hydrophobic graphene may interfere with insulin, causing partial denaturation, reducing the adipogenic differentiation of T-SCs.

Various MSCs and progenitor cells have been evaluated their differentiation capacity into the myogenic cells for skeletal muscle regeneration. MSCs derived from BM, adipose, and umbilical cord tissue, and their use in cell therapy to augment the skeletal muscle injury response, have also been reported[29-31]. An alternative cellular source for MSCs, T-SCs have been shown to differentiate into myogenic cells in vitro, and transplanting the myoblasts and myocytes generated from T-SCs mediates the recovery of muscle function following injury in vivo[32]. For myogenic differentiation, T-SCs are treated in three sequential steps: Sphere formation on a petri dish in low-glucose DMEM, rosette-like spread formation on a collagen-coated dish, and two weeks of myogenic induction. In this final step, the cells express myogenic markers, including desmin, dystrophin, MHC, skeletal markers, α-Actinin, TNNI1, and myogenin. Furthermore, the intramuscular injection of T-MSC-derived myogenic cells into myectomized C57BL/6 mice enhances muscle function as demonstrated by gait assessment and the restoration of the skeletal muscle structure.

Many clinical studies have indicated that BM-MSCs are safe and effective in the treatment of liver disease[33]. They can alleviate end-stage liver disease and improve symptoms and liver function[34,35]. However, some studies have indicated that BM-MSCs have the potential to aggravate fibrosis[36-38]. Thus, employing BM-MSCs as a therapy for liver fibrosis remains controversial.

As a novel cell source for treating liver disease, Park et al[39] demonstrated that T-SCs differentiate into hepatocyte-like cells and ameliorate live fibrosis via the activation of autophagy and the downregulation of TGF-β. A three-week culture in a differentiation medium containing IGF, HGF, dexamethasone, and oncostatin M led to the development of hepatocyte-like cells from T-SCs, as revealed by the expression of albumin and HNF-4α. In addition, transplanting T-SCs into a carbon tetrachloride (CCl4)-induced liver injury mouse model confirmed that T-SCs have a regenerative effect by migrating to the site of the liver injury and differentiating into hepatocyte-like cells. These results prove that T-SCs was able to differentiate into hepatocyte-like cells both in vitro and in vivo.

In addition to the direct differentiation of T-SCs into liver hepatocytes, T-SCs have also been investigated in terms of hepatogenic differentiation using PEG-L-PA thermogel[40,41]. The thermogel exhibited a physical modulus of 1000 Pa, which is similar to that of decellularized liver tissue. Three different 3D culture systems were compared in relation to the use of soluble factors such as hepatogenic growth factors. The spherical morphology and size of the encapsulated cells were maintained in the native 3D culture system during a culture period of 28 d, whereas the cells changed their morphology and aggregated significantly in 3D systems with growth factors. Hepatocyte-specific biomarker expression and metabolic functions were negligible in the native culture system. However, the expression levels of the hepatogenic genes of albumin and cytokeratin 18 and hepatocyte nuclear factor 4α were high in the two systems supplemented with growth factors. In addition, albumin and α-fetoprotein production were also significant[40]. PEG-L-PA thermogel thus provides a biocompatible microenvironment for the hepatogenic differentiation of T-SCs. In particular, the successful results of the growth factor encapsulated hydrogel system suggest that PEG-L-PA thermogel is a promising injectable tissue engineering system for liver tissue regeneration[41].

Metabolic disturbances associated with diabetes lead to a number of complications ranging from cardiovascular and cerebrovascular disease to neuropathy, retinopathy, nephropathy, and the poor healing of wounds[42]. The only curative therapy available is pancreatic islet cell replacement, for which suitable donors are rare and which requires immunosuppressant therapy to reduce rejection. Recently, stem cell therapy has been proposed for the treatment of diabetes. Transplanting insulin-secreting cells produced from various stem cells, including embryonic and induced pluripotent stem cells and MSCs has shown therapeutic effects in diabetic animals[43]. In addition, differentiating various MSCs, including BM-MSCs and adipose MSCs, into insulin-producing cells has been suggested[44]. Kim et al[10] investigated the efficiency of differentiating T-SCs into insulin-producing cells by comparing two different methods and found that T-SCs differentiated more efficiently with insulin-transferrin-selenium (ITS) than with β-mercaptoethanol. The ITS method is composed of three steps: Two days of culturing in high-glucose α-MEM with 1% fatty acid-free bovine serum albumin (BSA) and 1 × ITS on a nonadherent dish; four days of culturing in high glucose α-MEM with 1% fatty acid-free BSA, ITS, 3 mM taurine, and 10 mM nicotinamide; and four days of culturing in high-glucose α-MEM with 1% fatty acid-free BSA, ITS, 3 mmol/L taurine, 10 mmol/L nicotinamide, 100 nmol/L glucagon-like peptide, and 10 Nm exendin-4. Notably, T-SCs exhibited a differentiation capability that was superior to that of adipose cell-derived MSCs. Further, implanting T-MSC-derived insulin-producing cells significantly alleviated streptozotocin-induced glucose intolerance in mice. These results suggest that T-SCs have the potential to be reprogrammed into pancreatic β-cells and applied to the clinical treatment of diabetes in the future.

Hypoparathyroidism is a rare endocrine disorder, resulting in low serum calcium and increased serum phosphorus[45]. Hypoparathyroidism is the only hormonal insufficiency state that does not have a hormone-replacement-therapy approved. Current managements include supplementation with oral calcium and active vitamin D, which cause various life-long adverse effects[45].

Stem cells have shown some promise in treating hypoparathyroidism in clinical applications. It has been reported that human embryonic stem cells (hESCs) and differentiated thymic stromal cells can be used in the in vitro regeneration of parathyroid-like cells[46-48]. However, the use of hESCs have critical ethical limitations, and it takes over 10 weeks for thymic stromal cells to differentiate and secrete PTH. Because of this, T-SCs have been considered as an alternative cell source for cell therapy. Park et al[49] demonstrated that T-SCs differentiate into parathyroid-like cells that release intact PTH using the modified Bingham protocol. Briefly, T-SCs at 90% confluence were cultured in a differentiation medium containing activin A and soluble sonic hedgehog for 7-21 d. Surprisingly, the T-SC-derived parathyroid-like cells differentially secreted PTH in response to extracellular calcium levels. Further, the therapeutic effects of T-SC-derived parathyroid-like cells embedded in Matrigel in rats that have undergone a parathyroidectomy suggest that embedding differentiated T-SCs in hydrogel scaffolds is a promising strategy for restoring parathyroid function.