Background

Mesenchymal stem cells (MSCs) are adult stem cell populations derived from postnatal tissues. They retain fast proliferation rate, self-renewal capabilities, and lineage-specific differentiation potential. Moreover, MSCs are capable of producing growth factors, having anti-apoptosis properties, lacking carcinogenicity (Trounson and McDonald, 2015), and no adverse effects noted thus far in clinical trials (Lalu et al., 2012). Therefore, MSCs represent promising cell candidates for stem cell-based therapy and regenerative medicine (Tuan et al., 2003; Chen et al., 2008). Functional MSCs have been well established from various tissues, including bone marrow (BM) (Noth et al., 2002), adipose tissue (Zuk et al., 2001), peripheral blood (da Silva Meirelles et al., 2006), umbilical cord blood (Lee et al., 2004; Schuh et al., 2009), and other human organs (Campagnoli et al., 2001; Crisan et al., 2008). Among them, BM is regarded as the most accessible and reliable source for MSCs in adults, considering the wide distribution and well-characterized isolation methodology (Mezey et al., 2000). MSCs derived from human bone marrow (hBM-MSCs) have been clinically applied for a variety of injuries/disorders. Allogeneic hBM-MSCs transplantation-based tissue regeneration provides the possibility to re-establish bone-associated elements (Horwitz et al., 1999; Ghasroldasht et al., 2014). And autologous hBM-MSC transplantation could increase cartilage formation and repair the cartilage defects (Wakitani et al., 2007; Wu et al., 2013). However, the potential therapeutic applications of hBM-MSCs have been hindered by limited number of cell acquisitions, the limited cell long-term expandability in vitro, high cell heterogeneity and invasive procedure of cell isolation from human donors (Kretlow et al., 2008; Uccelli et al., 2008). Therefore, obtaining large quantities of functional MSCs from limited human donors is a challenge.

The large population of functional MSCs derived from human pluripotent stem cells (hPSCs) by in vitro differentiation, including human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs) could be an alternative to hBM-MSCs for therapeutic applications. MSCs derived from hESCs and hiPSCs (hESC-MSCs and hiPSC-MSCs) can overcome potential barriers for clinical therapeutic applications of hBM-MSCs, without capacity and function decline (Brown et al., 2014; Hynes et al., 2014). First, hPSCs represent infinitive cells ex vivo and in vivo, thus providing renewable cell sources for MSC generation with a high yield from various persons, without invasive procedure. Secondly, due to the reduced immune rejection and no tumorigenesis, MSCs derived from hPSCs can be used for autologous transplantation and personalized cell therapy (Peng et al., 2016; Gao et al., 2017). Thirdly, there has a great advantage to use isogenic hMSCs for autologous MSCs transplantation, in which the hPSCs can be genetically modified by endonuclease on patient PSCs with genetic mutations, and then differentiated into isogenic hMSCs. So far, hESCs are already used to generate clinically compliant MSCs, which displayed similar gene expression profiles and function as hBM-MSCs (Lian et al., 2007). Meanwhile, hiPSC-MSCs can significantly relieve hind-limb ischemia, in which hiPSC-MSCs were even more beneficial than hBM-MSCs due to superior survival and engraftment capability (Lian et al., 2010).

How to rapidly produce a large quantity of genetically identical MSCs from hPSC as a substitute for hBM-MSCs? Simple methods have been reported to effectively induce hPSCs into functional MSCs. Identical batches of hESC-MSCs can be reproducibly generated from hESCs in differentiation medium supplemented with 5 ng/ml FGF2 and 5 ng/ml PDGF-AB (Lian et al., 2007). Camilla and the colleagues successfully established mesenchymal progenitors from hESCs by single cell passage within high glucose medium and 10 ng/ml bFGF (Karlsson et al., 2009). MSCs can also be differentiated from hiPSCs in knockout medium containing 10 ng/ml bFGF, 10 ng/ml PDGF-AB, and 10 ng/ml epidermal growth factor (Lian et al., 2010). Furthermore, MSCs can be generated from hiPSCs via embryoid body formation with p38-MAPK inhibitor SB203580 treatment, in which hiPSC-MSCs could have less tumorigenicity (Wei et al., 2012). Our team previously developed a simple and efficient method to derive hiPSC-MSCs by culturing hiPSCs in low glucose medium for 2 weeks, and a serial passaging without embryoid body formation and flow sorting (Zou et al., 2013; Kang et al., 2014). However, a monolayer of feeder cells involved in the induction process has been a risk factor for therapeutic application (Kang et al., 2015). Another single step method to transform hESCs/hiPSCs into MSCs is to culture hESCs/hiPSCs in serum-free medium supplemented with TGF-β pathway inhibitor SB431542 for 10 days. This differentiation process was in an embryoid body independent and feeder cell-free manner. However, these hESC/hiPSC-MSCs showed limited adipogenic differentiation capacity, and this induction procedure was SB431542 dependent (Chen et al., 2012). In 2015, Fei Liu’s group also described a modified protocol to improve the differentiation of hiPSCs toward MSCs in mTESR1 medium supplemented with TGFβ inhibitor SB431542 in 7.5% CO₂ incubation, followed by repeatedly passaging cells without flow cytometric sorting. 88% of derived hiPSC-MSCs were positive for MSC surface markers, but the 7.5% CO₂ condition is not universal for cell culture, and SB431542 is needed for this induction process (Zhao et al., 2015).

Different methods for deriving functional MSCs from hPSCs have their own advantages, such as a simple protocol, or high efficiency. However, some current methods still have technical limitations regarding the MSCs production, including being time-consuming (more than 1.5 months), having less adipogenic differentiation capability of cells, and requiring additional chemicals or uncommon culture condition. Therefore, here, we updated a simple, efficient method for differentiating hESCs/iPSCs into MSCs. This protocol is free of embryoid body formation and feeder cell, approximately takes 28-30 days to generate high-quality MSCs. This system would provide a promising platform for generating a large scale of hiPSC-MSCs/hESC-MSCs from single donor for regenerative medicine.