Researchers have discovered an added layer of complexity in the network that determines human embryonic stem cell fate. A report publishing online April 30th in the journal Cell, a Cell Press publication, shows that a microRNA known as miR-145 lowers the activity of three key ingredients in the "recipe" for making embryonic stem cells. The discovery may have implications for improving the efficiency of methods designed to reprogram differentiated cells into embryonic stem cell-like cells and for the use of those transformed cells in replacing cells lost to disease or injury, the researchers said.
"The heart of the matter is that before this paper, we knew if you want to maintain a pluripotent state and allow self-renewal of embryonic stem cells, you have to sustain levels of transcription factors, including Oct4, Sox2 and Klf4," said Kenneth Kosik of The University of California, Santa Barbara. "We also knew that stem cells transition to a differentiated state when you downregulate those factors. Now we know how that happens a little better."
Transcription factors are genes that control other genes. On the other hand, microRNAs are single-stranded RNA molecules that control the activity of other genes. When microRNAs in the genome are transcribed from DNA, they target complementary messenger RNAs (mRNAs), which serve as the templates for proteins, to either encourage their degradation or prevent their translation into functional proteins. In general, one gene can be repressed by multiple microRNAs and one microRNA can repress multiple genes, the researchers explained. In a wide variety of developmental processes, microRNAs fine tune or restrict cellular identities by targeting important transcription factors or key pathways.
The new study adds embryonic stem cell identity to that list. Kosik's team found that levels of miR-145 change dramatically when human embryonic stem cells differentiate into other cell types. miR-145 was of particular interest because it had been predicted to target Oct4, Klf4 and Sox2. (Those three factors are perhaps best known as three of four ingredients originally shown to transform adult human skin cells into "induced pluripotent stem cells" (iPS cells), which behave in nearly every respect like true embryonic stem cells (http://www.eurekalert.org/pub_releases/2007-11/cp-srt111307.php). That four-ingredient recipe has since been pared down to one, Oct4, in the case of neural stem cells (http://www.eurekalert.org/pub_releases/2009-02/cp-sfc020209.php).)
A rise in miR-145 prevents human embryonic stem cells' self-renewal and lowers the activity of genes that lend stem cells the capacity to produce other cell types, the researchers report. It also sends the cells on a path toward differentiation. In contrast, when miR-145 is lost, the embryonic stem cells are prevented from differentiating as OCT4, SOX2 and KLF4 concentrations rise.
They also show that the control between miR-145 and the "reprogramming factors" goes both ways. The promoter for miR-145 is bound and repressed by OCT4, they found.
"It's a beautiful double negative feedback loop," Kosik said. "They control each other."
Because there is typically less "wiggle room" in the levels of microRNA compared to mRNA, further studies are needed to more precisely quantify the copy numbers of miR-145 and its targets to figure out exactly how this layer of control really works, Kosik said.
The findings in embryonic stem cells might also have importance for cancer.
"There are sets of microRNA that are widely up- or downregulated in cancers," he said, noting that several studies have specifically linked low miR-145 levels to various forms of cancer. "Tumor stem cells are the engines of tumors. If miR-145 is sustaining or maintaining a differentiated state, loss of that may have something to do with malignant transformation."
The researchers include Na Xu, The University of California, Santa Barbara, CA; Thales Papagiannakopoulos, The University of California, Santa Barbara, CA; Guangjin Pan, University of Wisconsin-Madison, Madison, WI; James A. Thomson2, Kenneth S. Kosik, The University of California, Santa Barbara, CA.
AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert! system.