Published online 11 October 2009 | Nature | doi:10.1038/news.2009.989


In search of true stem-like cells

Live-cell fluorescence imaging identifies bona fide reprogrammed cells.

Stem cellsFluorescence imaging could help resolve whether iPS cells have been properly programmed.Alamy

The next tools for reprogramming cells to an embryonic-like state might just be a camera and a set of fluorescently tagged antibodies.

Researchers imaged more than a million human cells in vitro as they changed from skin tissue cells, known as fibroblasts, into colonies of induced pluripotent stem (iPS) cells. As expected, many similar-looking colonies appeared, but only very few consisted of fully reprogrammed iPS cells. After assessing which were which, researchers led by Thorsten Schlaeger and George Daley of the Harvard Stem Cell Institute in Cambridge, Massachusetts went back and worked out how to predict which colonies would produce high-quality iPS cell lines by analyzing the images.

Robert Blelloch, who studies reprogramming at the University of California San Francisco, sees immediate practical applications. "It means that you can focus down on the most promising colonies and not assay everything." Currently, he says, many evaluations of techniques to boost reprogramming rates lump some partially reprogrammed cells together with fully reprogrammed ones. With better markers of pluripotency, he says, "you can look at the dish and count" and be more confident of your results.

False alarms

The analysis shows that individual markers for proteins such as alkaline phosphatase, SSEA-4, GDF-3, hTERT and NANOG that are sometimes used to assess reprogramming rates can be misleading. Instead, Schlaeger and colleagues identify a series of markers that can differentiate cells from similar-looking colonies into three types; two could form teratomas, a type of tumour, in immunocompromised mice — a standard assay for testing iPS cells, but only one of these shows epigenetic modifications indicating true iPS cells.

"I think these [kinds of] molecular criteria are going to be the best way to characterize iPS cells," says James Ellis, who directs the Ontario Human Induced Pluripotent Stem Cell Facility in Toronto. Lead author Schlaeger isn't so sure: "It's too early to tell if this could replace the teratoma assay." For now, he says, an assessment of how cells will behave is more informative than a collection of markers.

However, Schlaeger thinks that more researchers can use live imaging and antibodies to find more markers in reprogramming human iPS cells. When studying reprogramming in mouse cells, scientists routinely observe the expression of important genes by adding in tags that make the proteins produced by the genes to flouresce, and so allows them to see when a gene is active. This technique is much more difficult in human cells, so the Harvard researchers stained antibodies using a fluorescent dye instead. These fluorescently tagged antibodies could be added directly to growing cells to reveal the presence of particular molecules on cell surfaces. An automated microscope scanned and recorded images as the cells grew. "It's surprisingly easy," Schlaeger says. "It worked the first time we did it."

Humble beginnings

The antibodies and live cells weren't by themselves sufficient, however, to pinpoint truly reprogrammed colonies. Standard techniques for inducing cells to pluripotency rely on viruses to insert copies of four reprogramming genes into cells, and the researchers also tied expression of each of these genes to expression of green fluorescent protein. As cells reprogram, they turn on their own pluripotency machinery and silence the introduced genes; the subsequent dimming of GFP is an important marker of reprogramming. This means the live-image technique can't be used to evaluate newly reported reprogramming techniques that use protein factors and leave cells' genomes intact — an approach that may decrease risk of tumours and unpredictability.


The live-imaging approach may be of even more use in studying the beginning of reprogramming rather than the end product, says Schlaeger. "It allows us to identify earlier stages, so we can look at molecular events in these very rare cells." The technique allows researchers to watch just-emerging cell clusters and watch as markers turn on and off.

"It's extremely important to follow the fate of human iPS cells as they reprogram," says Ellis. "This is really the first paper to do that on single human iPS cell colonies induced using the standard retrovirus vector reprogramming approach." 

  • References

    1. Chan, E. M. et al. Nature Biotechnol. doi:10.1038/nbt.1580 (2009).
Commenting is now closed.