Testing pluripotency

A new open-access bioinformatics tool allows researchers to determine the pluripotency of newly derived cell lines by inputting microarray data

Jef Akst

| 2 min read

Researchers have developed a new, open-access tool for determining the ability of a cell line to differentiate into multiple tissue types, according to a study published online today in Nature Methods. Users need only to input their microarray data on the expression patterns of cells, and the program predicts their pluripotency.

Human embryonic stem cells
Image: Wikimedia commons,
Nissim Benvenisty

"A good 'transcriptome assay' can sometimes even detect differences that would otherwise be difficult to discern by morphology or immunostaining," which are used in other tests of pluripotency, stem cell biologist linkurl:Thorsten Schlaeger;http://stemcell.childrenshospital.org/about-us/leadership-faculty-staff/thorsten-m-schlaeger/ of the Children's Hospital Boston, who was not involved in the research, said in an email to The Scientist. "We will certainly give this a try."Pluripotency, the ability of a cell to generate any tissue in the body, is a hallmark of stem cells. But determining if a cell line is truly pluripotent has been somewhat of a challenge, as current techniques are cumbersome or inaccurate. The most common method for identifying pluripotency, for example, known as the teratoma assay, involves injecting the cells into immunodeficient mice and examining the tumors that develop. Not only is this assay time-consuming and technically challenging, a recent report suggests that both the methods and analysis of teratoma tests vary widely across researchers and cell lines."The teratoma assay is less than ideal for many reasons," Schlaeger said. "[It's] not standardized, only qualitative / semi-quantitative, slow, expensive, in vivo, [and] difficult to interpret."Recognizing the available techniques' shortcomings, stem cell biologist and bioinformatician Franz-Josef Müller of linkurl:Zentrum für Integrative Psychiatrie;http://www.zip-kiel.de/1willkommen/0willkommen.htm in Germany and his colleagues devised a bioinformatics-based program, dubbed PluriTest, that can analyze gene expression profiles to determine if a cell line is pluripotent."The idea is to take a lot of this super complicated bioinformatics and make it really useful and usable," Müller said. Analyzing 223 human embryonic stem cell and 41 induced pluripotent stem cell lines, the researchers' identified commonalities in gene expression to create a molecular definition of pluripotency. Inputting microarray data from a novel cell line, their model can then determine whether or not it is likely to be pluripotent, and whether it resembles any of the cell lines in the stored "bank" of cell lines profiles. The tool can even identify epigenetic or genetic abnormalities or unwanted differentiation."Transcriptome-based analysis can be very useful to distinguish cell and tissue types," Schlaeger said. However, he added, "I do not think this assay alone will be able to replace teratomas." Teratomas, he explained, still hold valuable information that this model may not be able to predict, such as whether a stem cell line has cells that are biased towards differentiating into one tissue type over another. "No single assay alone is likely to become a gold standard," he said.In addition to identifying pluripotent stem cells, Müller hopes to use the technology for other purposes, such as identifying disease states. "The power of the statistical model is that it can adapt to every type of data you can think of -- histone modifications, DNA methylation, everything that can fit into a matrix," he said. "It is a very general framework for assessing qualities in tissues and cells."F-J Müller, et al., "A bioinformatic assay for pluripotency in human cells," Nature Methods, AOP, doi:10.1038/nmeth.1580, 2011.
**__Related stories:__***linkurl:Pluripotency process unveiled;http://www.the-scientist.com/blog/display/56228/
[21st December 2009]*linkurl:How to ID human pluripotency;http://www.the-scientist.com/blog/display/56050/
[11th October 2009]*linkurl:Patient-ready iPS cells?;http://www.the-scientist.com/blog/display/55738/
[28th May 2009]