Teaching old dogs new tricks

25 February 2015
By Jian Yang

Epiblast stem cells (EpiSCs) and embryonic fibroblast cells (MEFs) can be reprogrammed to become naïve pluripotent stem cells.  Credit: Jian Yang

Epiblast stem cells (EpiSCs) and embryonic fibroblast cells (MEFs) can be reprogrammed to become naïve pluripotent stem cells. Credit: Jian Yang
Click to view large image

The same multifunctional small molecule is involved in embryo development, cancer therapy and now cell reprogramming. More secrets of this magic molecule will be revealed as we gradually understand its mechanism.

Retinoic acid is an important small molecule derived from Vitamin A, a fat-soluble vitamin that has multiple roles in development, growth and vision. During early embryo development, retinoic acid acts as an intercellular signalling molecule that helps to lay out the baby’s body plan.

Scientists at Cambridge University who wanted to study early embryo development first derived embryonic stem cell (ESC) lines in the 1980s from cells inside the embryos of young mice. These cells, known as inner cell mass, are the foundation of all three lineages and germ cells that make up an individual. Like inner cell mass, ESCs have two characteristics: self-renewal and pluripotency, the ability to differentiate to become almost any type of cell in the body.

Retinoic acid is one of the molecules widely used to investigate the mechanism of ESC differentiation. It can stimulate ESCs to differentiate into neuron cells, cardiac muscle cells and liver cells among others, by binding a retinoic acid receptor with a nuclear receptor known as retinoid X. This non-covalently bonded, two-factor structure (called a heterodimer) then binds to a specific DNA sequence to activate or repress gene expression.

Human embryonic stem cells were successfully derived in 1998. However, because of the scarcity of embryos and ethical concerns, it was difficult to study the molecular mechanism of early human embryo development.

Scientists tried to find alternative routes to develop human pluripotent stem cells. In 2006, Japanese scientist Shinya Yamanaka established a new type of stem cell by introducing four genes, Oct4, Sox2, c-Myc and Klf4 (4F) into differentiated cells to make them become ESC-like cells. He named these cells induced pluripotent stem cells (iPSCs). Because iPSCs are made from stem cells that have already differentiated, the whole genome of the cell must undergo dramatic epigenetic changes, called reprogramming. Until recently, iPSC technology has been widely used in biological mechanism and translational research.

Based on 4F reprogramming, we found another two genes (2F), Retinoic acid receptor gamma and Liver homolog 1. Used together with 4F, these genes can rapidly and efficiently reprogram both mouse and human cells to become iPSCs.

To apply iPSCs in regenerative medicine, the cells must be made without allowing any foreign genetic material to become part of the genome. This is a common occurrence when cells are reprogrammed using another animal or human cell. We developed a technique that doesn’t need a second cell; instead, a piece of genetic material, called an episome, is placed directly into the cell, delivering 6F (4F+2F). These six genes efficiently reprogramme the genome to produce an error-free iPSC.

The next question is, can we improve reprogramming efficiency by manipulating retinoic acid signalling? To answer this, we took advantage of our system to activate signalling with different retinoids, chemical compounds also derived from Vitamin A, or inhibit signalling with retinoid antagonists.

Interestingly, we found that only a narrow range of low-concentrations of retinoids promoted reprogramming significantly. Importantly, the concentration of retinoic acid must be at least 100 times lower than the concentration used for ESC differentiation in mice.

We tested this technique on fibroblasts, cells found in connective tissues, and epiblast stem cells (EpiSCs), found in early embryos. Because EpiSCs are already pluripotent but with limited differentiation potential, the concentration of retinoic acid required is lower than that used for fibroblast reprogramming. We concluded that, to make use of retinoic acid signalling in reprogramming, its activation should be kept at an optimal level, which is cell-type dependent.

We’re beginning to understand retinoic acid’s magic qualities and we hope that this unassuming small molecule may soon have the potential to make cell-reprogramming even more efficient and accurate, speeding the development of regenerative treatment.

Jian Yang is a Postdoctoral Fellow in Pentao Liu’s group at the Wellcome Trust Sanger Institute.


  • Yang J, et al (2015). Signalling through retinoic acid receptors is required for reprogramming of both MEFs and EpiSCs to iPSCs. Stem CellsDOI:10.1002/stem.1926

Related Links: