Congress is about to take on the issue of embryonic stem (ES) cells again. President Bush has already vetoed a bill regarding federal funding of research on stem cells that involve creating new lines of ES cells because it destroys the embryo in the process. The US is falling behind other countries with regard to our research in this area. Now, Shinya Yamanaka from Kyoto University in Japan, has discovered a method to transform mature skin cells in the mouse into a cell type that is indistinguishable from ES cells. These cells exhibit all the hallmarks of pluripotent stem cells; that is to say that they have been shown to have the ability to become all forms of tissue including reproductive cells.
The method used by Yamakana and two other independent labs relies on properties of retroviruses. These viruses actually insert their own DNA into the host’s genome upon invading a cell. If you added genes to a virus like this, those genes would also be inserted into the host’s genome. Yamakana’s lab infected mouse cells with four such engineered viruses, each one containing a different gene. The success rate is so far just one in 1000. This may be due to needing more factors or an artifact of infection by the retroviruses (more on retroviruses below). The surviving cells were tested to compare their traits with those of stem cells derived from embryonic sources. They were found to possess all the traits of such cells.
Three of the genes that were inserted into the differentiated cells to transform them into pluripotent stem cells (Oct4, Sox2, and c-Myc) had been known to stem cell scientists as playing a role in pluripotency. The fourth (Klf4) was a new addition. All four of the genes are transcription factors. A transcription factor is a type of control signal for other genes. These genes code for a small protein that binds to DNA near the start of some other gene to either activate or inhibit the other gene (watch movie of how a gene gets transcribed). One can imagine transcription factors that target many genes at once. In this way a cell can activate or deactivate entire cellular programs by at first only turning on a single gene. It is not surprising then that the genes used here turn out to be transcription factors!
Oct4 is a gene that is originally provided by the mother in the egg cell before fertilization. It has been found that either increasing or decreasing the expression of this gene will trigger the cell to differentiate into specialized cell types ending their career as stem cells.
Sox2 has been implicated in differentiation of neurons. It may also control multiple other transcription factors down the line as well.
c-Myc is an interesting one. It binds certain areas of DNA and recruits a group of enzymes that can modify proteins called histones. This is another control mechanism that the cell uses to decide which areas of DNA are allowed to be activated. When in a resting cell, the DNA is actually wrapped around these histones like a ball of string. This causes much of the DNA to be hidden, just like you can only access the outer layer of the ball of string. c-Myc can tell the histones to unravel the DNA so that the hidden sections can be accessed and genes in those regions can be turned on. It is also known to influence genes that regulate cell division (the cyclins and p21). Interestingly, c-Myc has been found in an over active state in many cancer cells. Cancer cells and stem cells actually have similar characteristics in some regards, so it is not too surprising, albeit a bit unsettling, to find that a gene be involved in both.
Like c-Myc, Klf4 controls genes involved with controlling cell division. One gene that it directly inhibits is called p53. p53 is a gene that polices the cell for DNA damage. As the cell gets ready to divide it must synthesize a new copy of its DNA for the second cell to receive. If the DNA is damaged when it is duplicated, the damage can cause mistakes in the sequence of the new cell’s version of the genome. If this happened, then every cell that the new cell gave rise to would also have the mistake. When there is DNA damage, p53 holds the cell in the stage right before DNA duplication so that repair enzymes can have a chance to fix the problem. If the problem is fixed, the cell is allowed to proceed. If the damage persists, p53 sets off a series of changes in the cell that ultimately lead to the cell’s suicide. It is better that the cell dies with bad DNA than pass genes with mistakes on to generations of new cells. This is why p53 has been given the evocative title of “the guardian angel gene”. More than half of human cancers involve a deletion or mutation of p53. If a person is born with only one functional copy of p53, they are virtually certain to develop some type of tumor by early adulthood. Because Klf4 suppresses p53 it must be regulated very well so as not to affect p53’s very important function.
As I see it the real value in this technique is that it promises to allow MANY more labs to study stem cells and the challenges to controlling them; however, we must not over estimate it’s therapeutic usefulness at this time. I do not see this as a way to get around the embryonic stem cell problem entirely. It comes with a very dangerous side effect in its current form. 20 percent of the mice in Yamakana’s study developed tumors. The cancer danger is two fold.
The studies used a retrovirus delivery system to insert the new genes into the cell’s own genome. The problem with this method, is that we have no control over where the genes are placed in the genome. Hopefully they are inserted nicely into a region of DNA between genes. However, the chances are also very good that they will be inserted right in the middle of a very important gene. This will effectively delete that gene from the cell’s genome, as well as those of ALL cells derived from this cell. All organs made from that cell will not have that gene. This is a big problem if the deleted gene is something like p53, that is part of our defense against cancer. To fix this problem we need to come up with a way to insert the genes into a region that we can choose. There are other delivery systems, or vectors, that can be tried. In some animals including insects we have managed to achieve site-specific insertions and may be able to try modified versions of these techniques.
The second source of cancer risk arises from the genes themselves. Rlf4 reduces the effect of p53, and c-Myc is also strongly correlated to tumors. Here we must work on how to structure the control of these genes so that they will be active when we want them to but quiet down later. Work is being done in this area, and in fact, Yamakana’s method now provides a way for more people to study this exact problem.
This development will not yield therapeutic benefits for some time. It has not even been attempted in human cells as of yet. Further more, its cancer risk, as well as its non-specific retrovirus delivery system, will probably prohibit the current version’s use in human therapies. This IS however an extremely important achievement. While it may not be useful to cure any sicknesses, its real value is that it will allow many more labs to study the challenges of controlling stem cells. This method will allow even modestly equip labs to produce and study stem cells that act in all currently known ways exactly like stem cells derived from embryos. This may effectively allow us to claw our way back into the stem cell research race. Mind you I am not saying that we don’t need to be able to study ES cells. These manufactured pluripotent cells behave like ES cells as far as we can tell, but we may be missing something. It remains to be seen whether Yamakana stem cells will yield discoveries that will be applicable to all stem cells, but for now it looks as though we have a very valuable new tool in our belt. I can’t wait to see what we learn with these!