(Image from Takahashi et al., Induction of Pluripotent Cell form Adult Human Fibroblasts by Defined Factors, Cell (2007), doi:10.1016/j.cell.2007.11.019)

Ok. Now I believe.

Two groups working independently–Dr. Yamanaka’s lab in Japan and Dr. Thomson’s Lab in Wisconsin–have converted human cells into embryonic stem cell-like cells. This tremendous accomplishment is on par with the initial creation of human embryonic stem cells about ten years ago, the completion of the human genome project and development of gene knockdown technology.

With this trick, skin cells can be converted into embryonic stem cell-like cells that can become any cell type in the body, including difficult to acquire human heart and brain cells.

I’ve written about this technique rather skeptically in the past. All of the previous work was with mouse cells. Now with two groups independently showing it works also with human cells, I’m pretty convinced. Yes, these two papers are pretty sloppy. The Thompson one reads like it was written over the weekend and the Yamanaka paper has a few glaring flaws. Still, the evidence has tipped in this case.

Embryonic stem cells–created from a few hundred cells in a very early human embryo–have the ability to become any cell type in the body, including heart and brain cells. Adult cells, like those found in your skin, are restricted to doing only a few things at a time.

After adding four genes–Oct4, SOX2, cMyc, and Klf4 according to Yakanama; Oct4, SOX2, NANOG and LIN28 per Thompson–to differentiated cells and growing the cells in careful culture conditions, some of the cells in the culture become quite like embryonic stem cells. Both groups went on to show these “Induced pluripotent cells” (iPS cells) can be turned into useful cell types just like embryonic stem cells. For example, the iPS cells were efficiently turned into beating human heart cells–all in a dish.

The technique still isn’t perfect. The four genes are still being added with retroviruses that permanently change the cell’s genetic material. Additionally, the proper differentiation of the cells requires these added genes to be silenced by a process that is still not entirely understood. Finally, at least one of the genes, Oct4, is cancer-causing. Therefore, this new technique is more useful for the lab than the clinic. None of these problems are unsolvable.

Why are these iPS cells so exciting?

The ability to make patient- or disease-specific stem cells should give a boost our unraveling of how complex combinations of genes cause illness or change responses to medications–particularly in combination with the human genome project and the ability to selectively shut off genes in cultured cells through RNA interference.

Let’s say you have two patients, one who responds to a cardiac drug and another who doesn’t. If you want to figure out why, you could now pluck some skin cells from each patient, and create iPS cells from each. In turn, iPS cells can be differentiated into heart cells (or blood vessel cells, or neurons, or liver cells and so on.) The heart cells can then be tested with the drug in a dish. Genetic differences between the two sets of heart cells–both in what genes each patient has, as well as which are on in their heart cells–can be determined and compared to the reference human genome sequence. The likely candidate genes can be turned off to figure out which ones are required.

Right now, iPS cells would be too dangerous to treat patients. Still, with a few relatively straightforward improvements, iPS cells could easily be a new cell source for treatments.Thus far, the techniques developed for embryonic stem cells–how to turn them into brain or heart cells–seem to work for iPS cells. That’s fantastic news. New heart cells made from iPS cells–all from a skin stem cell plucked from the patient–could replace the billion or so lost in a heart attack, for example.

The clinical medicine stuff is pure speculation. The impact on modern molecular biology of this technique–if it hold true; please hold true–is hard to understate. I’m excited. You should be too.