In 2000, scientists began identifying and characterizing proteins that help drive cells into the unique properties of embryonic stem cells. In August 2006, Shinya Yamanaka and colleagues at Kyoto University used four of these genes, all of which encoded transcription factors, Oct3/4, Sox2, c-Myc, and Klf4, to reprogram mouse skin cells to stem cells that exhibited most but not all of the properties of embryonic stem cells. The following year, Yamanaka’s team and James Thompson’s team at the University of Wisconsin–Madison concomitantly made induced pluripotent stem cells using Oct3/4, Sox2, Lin28, and Nanog.
Since this time, researchers have defined the function of these genes in detail in mouse embryonic stem cells (mESCs). Underneath this research was the assumption that the function of these genes in mESCs was the same in human cells. New research, however, has seriously challenged this notion.
Natalia Ivanova and her colleagues at Yale University in New Haven, CT, planned to study other genes that might be involved in maintaining pluripotency in human ESCs (hESCs). They used gene silencing to define the function of individual genes in hESCs. When they first silenced three of the genes that encode traditional pluripotency factors (Nanog, Oct4, and Sox2) they were very surprised to find that the stem cells did not act as they had expected. In mouse ESCs, Nanog, Oct4, and Sox2 bind the same locations on the genome and act cooperatively to maintain stem cell self-renewal and pluripotency. When any one of those factors is unable to perform its function, mESCs differentiate into extraembryonic tissues, such as placenta.
However, when they silenced each of the three factors in three different hESC lines, Ivanova and her team identified fundamentally different roles for these proteins in hESCs. First, the three factors prevent hESCs from transforming into non-embryonic tissues. Specifically, Nanog appears to prevent cells from becoming neuroectoderm, which is a tissue that eventually becomes the nervous system. Sox2 prevents cells from becoming mesoderm, which forms connective tissue and muscle.
Oct4 has varying roles depending on the presence of a protein called BMP4. In the absence of BMP4, inactivated Oct4 induces ectoderm, the outer layer of an embryo that forms the nervous system and skin, but in the presence of BMP4, it specifies extraembryonic cell fates. Also, Ivanova found that the trio does not function as a complex. Instead, Sox2 is the “odd man out,” since silencing Sox2 does not prevent hESCs from maintaining pluripotency in hESCs since Sox3, a related protein, compensates for Sox2’s absence. “It just shows you that in human [ESCs], repression by these three factors works in a completely different way” than in mouse ESCs, said Ivanova.
The findings could have a major impact on embryonic stem cell research. Mouse ESCs are traditionally easier to grow in culture than hESCs. One can grow sufficient mESCs for use in an experiment in two days, while it may take two months to grow enough hESCs. By understanding how human ESCs are regulated by these factors may help scientists fine-tune and speed up the expansion of hESCs in culture.
Also, because Nanog and Oct4 appear to be involved in the differentiation of the ectoderm, scientists may be able to use this knowledge to come up with new ways to inhibit these factors to improve differentiation of hESCs into neurons, which are quite valuable in a number of medical and scientific applications.
For their next project, the Yale team plans to get back to their original investigation of additional factors involved in hESC pluripotency. “Some other genes may be contributing to the regulation of self-renewal and differentiation,” said Ivanova. “We’re going to try to look at what these other players might be, to find out what else regulates this process.”