However PSCs have major safety concerns such as teratoma
However, PSCs have major safety concerns, such as teratoma formation due to possible contamination of the residual undifferentiated PSCs (Lee et al., 2013a). Teratoma is a unique characteristic of PSCs owing to their pluripotency and ability to undergo unlimited proliferation (Gruen and Grabel, 2006). These properties (e.g., pluripotency and limitless proliferation), which theoretically allow the sufficient supply of cardiomyocytes, cardiac progenitors, and endothelial cells, make PSCs the most promising cell source for regenerative medicine (Robinton and Daley, 2012). The unlimited proliferation capacity of PSCs is similar to the proliferation of cancer cells, which results from high telomerase activity (Thomson et al., 1998) and constant inhibition of retinoblastoma protein (Burdon et al., 2002). Therefore, the presence of a few undifferentiated PSCs or incompletely differentiated purchase Rosiglitazone that escape the strict sorting process would be a high risk factor for tumor development in safe stem cell therapies (Masuda et al., 2014). However, teratoma or tumor formation following xenotransplantation of human (h)PSCs into animal models (mostly rodents) has been barely reported (Kriks et al., 2011; Laflamme et al., 2007), with a few exceptions (Brederlau et al., 2006; Seminatore et al., 2010). A recent study by Doi et al. (2012) demonstrated that extended differentiation of hPSCs reduces the risk of teratoma formation in a primate model. Therefore, stringent selection of differentiated cells by flow cytometry for selective sorting (Cho et al., 2007) or prolonged differentiation (Brederlau et al., 2006; Zhang et al., 2001) would be sufficient to avoid tumorigenicity. Nevertheless, considering the “host-dependent tumorigenicity” of ESCs (e.g., xenotransplantation of hPSCs into a rodent model) (Erdö et al., 2003), the risk of teratoma formation by hPSCs in humans remains to be resolved, especially given that several clinical trials of hPSC-based therapies are ongoing (Cyranoski, 2013). Not only mouse embryonic bodies (mEBs), but also differentiated cells from mouse embryonic stem cells (mESCs) can reportedly form teratomas when transplanted into a rodent model, even after stringent cell sorting (Arnhold et al., 2004; Fujikawa et al., 2005; Moon et al., 2013). A number of approaches have been developed to reduce the risk of teratoma formation, including antibody-based selective elimination (Tang et al., 2011), small molecules (Ben-David et al., 2013; Lee et al., 2013b), and integration of a suicide gene (Li and Xiang, 2013). Among them, the use of various chemicals to achieve selective cell death of PSCs (referred as “stemotoxic” in a review, see Knoepfler, 2009) may have unexpected caveats, despite being highly efficient (Knoepfler, 2009). Additionally, there is no evidence indicating that such chemical treatment ensures functionality of differentiated cells in vivo (Masuda et al., 2014). Therefore, we attempted to eliminate undifferentiated PSCs without using small molecules and reveal that the differentiated cells after selective ablation of undifferentiated PSCs remain fully functional in vivo. To this end, KillerRed (KR), an artificial photosensitizer protein derived from a hydrozoan chromoprotein (Bulina et al., 2006) which produces reactive oxygen species (ROS) when exposed to visible light of 540–580 nm (Wang et al., 2012), was introduced into PSCs with a pluripotent-specific promoter such that it was only expressed in undifferentiated PSCs. A single treatment with visible light successfully eliminated KR-expressing mouse PSCs (KR-mPSCs) in an ROS-dependent manner, while endothelial cells (ECs) differentiated from KR-mPSCs survived and remained functional in vitro and in vivo. The similar results were repeated in a human ESC (hESC) model. Importantly, KR-mPSCs did not form teratomas after being exposed to the light, whereas ECs differentiated from KR-mPSCs effectively regenerated blood vessels under ischemic conditions in a rodent model. Our results strongly support and provide the “proof of concept” that a KR-based suicide gene approach in which visible light selectively induces phototoxicity in PSCs would be an efficient strategy to reduce the risk of tumorigenicity in PSC-based cardiovascular repair.