The team of Liu et al. generated endoderm-derived human induced pluripotent stem (iPS) cells from primary hepatocytes. 1 However, they generated human iPS cells by using viral transgenes. 1 Clinical applications of human iPS cells require avoiding viral transgenes. On the other hand, the reprogramming of human cells with only small molecules has yet to be reported. Therefore, we tried to reprogram human liver progenitor cells with only two small molecules. 2 First, octamer3/4-positive human liver progenitor cells were treated with 2 0 O-methyl-microRNA-145 as microRNA-145 inhibitor (100 nmol/L for 96 hours; after that, 50 nmol/L for 72 hours) and transforming growth factor-beta ligand (100 pM for 48 hours) in human embryonic stem (hES) cell medium. 2 As a result, we could generate human iPS cells from human liver progenitor cells only by use of small molecules. 2 The human iPS cells were similar to hES cells in morphology, proliferation, surface antigens, gene expression, and epigenetic status of pluripotent cell-specific genes. 2 Furthermore, these cells could differentiate into cell types of the three germ layers in vitro and in teratomas. 2 Therefore, we designated the human iPS cells as chemicals-human induced pluripotent stem (ChiPS) cells. 2 On the other hand, although Liu et al. did not show the risk evaluation of malignant transformations for the human iPS cells lines that they generated, 1 we performed the risk evaluation. 2 It was reported that cancer risk for patients with Down syndrome was less than healthy individuals, and the microvessel density (MVD) within severe combined immunodeficient (SCID) mice in which human iPS cells derived from patients with Down syndrome were transplanted was also less than the MVD in SCID mice in which human iPS cells derived from healthy people were transplanted. 3 Therefore, according to the method of Baek et al., 3 by using MVD within SCID mice in which ChiPS cell lines as human iPS cell lines were transplanted, we performed the risk evaluation of malignant transformations for the cell lines. As a result, the MVD in our study 2 was equal to the case 3 of patients with Down syndrome.
Furthermore, we tried to differentiate human normal hepatocytes from ChiPS cells as human iPS cells, according to the method of Liu et al. 1 As a result, we could generate mature hepatocytes 21 days after the initiation of differentiation (Fig. 1). Moreover, according to the method of Liu et al., 1 although we evaluated cytochrome P450 (CYP450) metabolism in ChiPS cell-derived mature hepatocytes, the CYP3A4 and CYP1A2 activity appeared to be the same as in the case of the ihH10 cell line that Liu et al. 1 generated.
In conclusion, human iPS cells that Liu et al. 1 or we 2 generated would be useful for the study of liver disease pathogenesis. However, our ChiPS cells 2 would have an advantage in clinical applications of human iPS cells.