Although iPS cells can be used to make any type of cell in the body, certain types have proven more difficult than others. In particular, organs such as the lung, liver and pancreas, which all originate from the endoderm, have proven problematic. One reason, according to Professor Kenji Osafune, is our poor understanding of endoderm development. "It is unclear if definitive endoderm cells have the same differentiation propensity," he explains. If not, then differentiation protocols using definitive endoderm cells would be ineffective on a subpopulation, which could explain the difficulty in generating the desired cell type. A new study from his lab shows that indeed, different developmental stages of definitive endoderm produce progenitors for different cell types. This finding not only has important implications on development, but also stem cell-based therapies.

Until now, the different stages of definitive endoderm had only been postulated, but never observed. Osafune's team, however, took advantage of CHIR99021, a small molecule that was previously used to show the same property is true for mesoderm. By simply changing the concentration of this drug in the differentiation protocol, the researchers were able to show that definitive endoderm derived from human iPS cells could be divided into anterior and posterior definitive endoderm, which resulted in different progenitor cells. The anterior definitive endoderm could be broken further into two groups by adding a second small molecule to the protocol. These two groups could then be differentiated into distinct cell populations that respectively resembled lung and liver cells.

Osafune is excited about the basic science impact this finding should have, saying, "This is a good model to study endoderm development." Moreover, the discovery that endoderm can be divided into different populations of progenitors should also be a key step in preparing different endoderm-lineage cells in the lab for patient care.

• doi: 10.1016/j.diff.2016.04.002