Research Highlights

Nature Reports Stem Cells
Published online: 6 March 2008 | doi:10.1038/stemcells.2008.44

Reprogramming kinetics

Monya Baker¹

Virally introduced genes must be active for several days to induce pluripotency

Multiple labs have now shown that adult cells can be reprogrammed into an embryonic-like state¹. A suite of virally introduced genes is necessary to kick-start the reprogramming process, but gradually the cell's own endogenous pluripotency genes become active, and the viral genes are silenced. Differentiated cells take weeks to become pluripotent, and some researchers estimate that fewer than 1 in 500 cells that take in copies of each gene reach the embryonic-like state. Two papers in Cell Stem Cell dissect events in the reprogramming process, and their insights could help researchers attempt new reprogramming approaches that do not use viruses.

The research teams, one led by Rudolf Jaenisch at the Massachusetts Institute of Technology in Cambridge, Massachusetts, and the other by Konrad Hochedlinger at Massachusetts General Hospital in Boston, both started with cultured mouse skin cells (fibroblasts) and added viruses carrying the four genes originally established as required for inducing pluripotency (Oct4, Sox2, Klf4 and c-Myc). Both teams had experience reprogramming fibroblasts with this method before. This time, however, they used versions of the genes that could only turn on in the presence of the small molecule doxycycline. That way, they could tell how long the reprogramming cells needed the transgenes to stay active.

Mouse skin cells reprogrammed with inducible viruses can contribute to other mouse tissues; in culture, they can help show how reprogramming works

Konrad Hochedlinger

Hochedlinger's team looked for the initiation of several events necessary for pluripotency: downregulation of genes normally active in fibroblasts, reactivation of the X chromosome, plus the upregulation of proteins that are characteristic of embryonic stem (ES) cells, such as telomerase, Sox2, Oct4 and stage-specific embryonic antigen-1 (SSEA-1). They found that the fibroblast genes were shut down quickly, in as little as two or three days². Cells first started expressing the gene for SSEA-1 at around five days. The other necessary events happened later. At each step, the researchers could sort cells according to the presence of particular pluripotency markers (or the absence of differentiation markers) and get a higher percentage of cells that would go on to produce induced pluripotent stem cells. "This now allows [us] to zoom into these intermediate cell populations and ask what's going on at the molecular level," says Hochedlinger.

Jaenisch's team first showed that endogenous pluripotency genes must become active in a particular sequence³, a finding that is consistent with other studies. For this, they used cell-sorting techniques, plus mouse cell lines engineered so that green fluorescent protein would be produced when one of two endogenous pluripotency genes (Nanog or Oct4) was expressed. These genes actually turn on late in the process. The first gene to activate was that encoding alkaline phosphatase, at around three days; the next was that for SSEA-1, at around nine days. Nanog and endogenous Oct4 were detectable by cell-sorting machines at around 16 days. Even 21 days after culture, transgene activity greatly boosted reprogramming rates, though the rates were still low. Although the cells looked morphologically different as early as three days after infection, they always reverted back to fibroblasts if doxycycline was removed from the culture within two weeks of the delivery of the transgenes.

Hochedlinger's team showed that the transgenes were necessary for about ten days, after which the cells' endogenous machinery sustained pluripotency. To study the kinetics of transgene silencing, they infected cells with a retrovirus coding for red fluorescent protein so that they could easily tell if the retrovirus was active or silent. Silencing began to occur as early as three days after infection. In colonies of cells examined 13 days after transfection, the researchers found that cells expressing red fluorescent protein (meaning the retrovirus had not been silenced) never expressed a green fluorescent protein tied to expression of pluripotency gene Sox2, and vice versa. This suggests that full reactivation of the pluripotency genes requires silencing of the viruses.

Jaenisch's team created versions of the transgenes that could not be silenced, and found that although transformed cells took on properties that are characteristic of ES cells, they did not differentiate when injected into mice, unlike ES cells or other induced pluripotent stem cells.

Next steps include exploring the roles and relationships of different transgenes and trying to figure out why reprogramming cells through this method seems to take so much more time than reprogramming through nuclear transfer. "Many sequential events have to happen. We have to re-establish the core circuitry of pluripotency," says Jaenisch.

Author affiliation

1. Monya Baker is news editor of Nature Reports Stem Cells.