STAP Cells: The Plot Thickens Even More


You might remember that Charles Vacanti and researchers at the RIKEN Institute in Japan reported a protocol for reprogramming mature mouse cells into pluripotent stem cells that could not only integrate into mouse embryos, but could also contribute to the formation of the placenta. To convert mature cells into pluripotent cells, Vacanti and others exposed the cells to slightly acidic conditions or other types of stressful conditions and the cells reverted to a pluripotent state.

Even though Vacanti and others published these results in the prestigious journal Nature, as other scientists tried to replicate the results in these papers, they found themselves growing more and more frustrated. Also, some gaffes with a few of the figures contributed to a kind of pall that has hung over this research in general.

The original makers of these cells, stress-acquired acquisition of pluripotency or STAP cells, have now made a detailed protocol of how they made their STAP cells publicly available at the Nature Protocol Exchange. Already. it is clear that a few things about the original paper are generating many questions.

First of all, Charles Vacanti’s name does not appear on the protocol. He was the corresponding author of the original paper. Therefore the absence of his name raises some eyebrows. Secondly, the authors seem to have backed off a few of their original claims.

For example one of the statements toward the beginning of the protocol says, “Despite its seeming simplicity, this procedure requires special care in cell handling and culture conditions, as well as in the choice of the starting cell population.” Whereas the original paper, on the first reading at least, seemed to convey that making STAP cells was fairly straightforward, this seems to no longer be the case, if the words of this protocol are taken at face value.

Also, the protocol notes that cultured cells do not work with their protocol. The authors write, “Primary cells should be used. We have found that it is difficult to reprogram mouse embryonic fibroblasts (MEF) that have been expanded in vitro, while fresh MEF are competent.”  This would probably explain inability of several well-regarded stem cell laboratories to recapitulate this work, since the majority of them probably used cultured cells. This, however, seems to contradict claims made in the original paper that multiple, distinct cell types could be converted into STAP cells.

Another clarification that the protocol provides that was not made clear in the original paper is that STAP cells and STAP stem cells are not the same thing. According to the authors, the protocol provided at Nature Protocol Exchange produces STAP cells, which have the capacity to contribute to the embryo and the placenta. On the other hand, STAP stem cells, are made from STAP cells by growing them in ACTH-containing medium on feeder cells, after which the cells are switched to ESC media with 20% Fetal Bovine Serum. STAP stem cells have lost the ability to contribute to extra-embryonic tissues.

Of even greater concern is a point raised by Paul Knoepfler at UC Davis. Knoepfler noticed that the original paper argued that some of their STAP cells were made from mature T cells. T cells rearrange the genes that encode the T cell receptor. If these mature T cells were used to make STAP cells, then they should have rearranged T cell receptor genes. The paper by Vacanti and others shows precisely that in a figure labeled 1i. However, in the protocol, the authors state that their STAP cells were NOT made from T-cells. In Knoepfler’s words: “On a simple level to me this new statement seems like a red flag.”

Other comments from Knoepfler’s blog noted that the protocol does not work on mice older than one week old. Indeed, the protocol itself clearly states that “Cells from mice older than one week showed very poor reprogramming efficiency under the current protocol. Cells from male animals showed higher efficiency than those from female.”  Thus the universe of cells that can be converted into STAP cells seems to have contracted by quite a bit.

From all this it seems very likely that the STAP paper will need to go through several corrections. Some think that the paper should be retracted altogether. I think I agree with Knoepfler and we should take a “wait and see” approach. If some scientists can get this protocol to work, then great. But even then, multiple corrections to the original paper will need to be submitted. Also, the usefulness of these procedure for regenerative medicine seems suspect, at least at the moment. The cells types that can be reprogrammed with this protocol are simply too few for practical use. Also, to date, we only have Vacanti’s word that this protocol works on human cells. Forgive me, but given the gaffes associated with this present paper, that’s not terribly reassuring.

Spiking Stem Cells to Generate Myelin


Regenerating damaged nerve tissue represents a unique challenge for regenerative medicine. Nevertheless, some experiments have shown that it is possible to regenerate the myelin sheath that surrounds particular nerves.

Myelin is a fatty, insulating sheath that surrounds particular nerves and accelerates the transmission of nerve impulses. The myelin sheath also helps neurons survive, and the myelin sheath is attacked and removed in multiple sclerosis, a genetic disease called Charcot-Marie-Tooth disease, and spinal cord injuries. Being able to regenerate the myelin sheath is an essential goal of regenerative medicine.

Fortunately, a new study from a team of UC Davis (my alma mater) scientists have brought this goal one step closer. Wenbig Deng, principal investigator of this study and associate professor of biochemistry and molecular medicine, said, “Our findings represent an important conceptual advance in stem cell research. We have bioengineered the first generation of myelin-producing cells with superior regenerative capacity.”

The brain contains two main cell types; neurons and glial cells. Neurons make and transmit nerve impulses whereas glial cells support, nourish and protect neurons. One particular subtype of glial cells, oligodendrocytes, make the myelin sheath that surrounds the axons of many neurons. Deng and his group developed a novel protocol to induce embryonic stem cells (ESCs) to differentiate into oligodendrocyte precursor cells or OPCs. Even though other researchers have made oligodenrocytes from ESCs, Deng’s method results in purer populations of OPCs than any other available method.

Making OPCs from ESCs is one thing, but can these laboratory OPCs do everything native can do? When Deng and his team tested the electrophysiological properties of their laboratory-made OPCs, they discovered that their cells lacked an important component; they did not express sodium channels. When the lab-made OPCs were genetically engineered to express sodium channels, they generated the characteristic electrical spikes that are common to native OPCs. According to Deng, this is the first time anyone has made OPCs in the laboratory with spiking properties. Is this significant?

Deng and his colleagues compared the spiking OPCs to non-spiking OPCs in the laboratory. Not only did the spiking OPCs communicate with neurons, but they also did a better job of maturing into oligodentrocytes.

Transplantation of these two OPC populations into the spinal cord and brains of mice that are genetically unable to produce myelin also showed differences. Both types of OPCs were able to mature into oligodendrocytes and produce myelin sheaths, but only the spiking OPCs had the ability to produce longer and thicker myelin sheaths.

Said Deng, “We actually developed ‘super cells’ with an even greater capacity to spike than natural cells. This appears to give them an edge for maturing into oligodendrocytes and producing better myelin.

Human neural tissue has a poor capacity to regenerate and even though OPCs are present, they do not regenerate tissue effectively when disease or injury damages the myelin sheath. Deng believes that replacing glial cells with the enhanced spiking OPCs to treat injuries and diseases has the potential to be a better strategy than replacing neurons, since neurons are so problematic to work with in the laboratory. Instead providing the proper structure and environment for neurons to live might be the best approach to regenerate healthy neural tissue. Deng also said that many diverse conditions that have not been traditionally considered to be myelin-based diseases (schizophrenia, epilepsy, and amyotrophic lateral sclerosis) are actually now recognized to involve defective myelin.

On that one, I think Deng is dreaming. ALS is caused by the death of motor neurons due to mechanisms that are intrinsic to the neurons themselves. Giving them all the myelin in the world in not going to help them. Also, OPCs made from ESCs will be rejected out of hand by the immune system if they are used to regenerate myelin in the peripheral nervous system. The only hope is to keep them in the central nervous system, but even there, any immune response in the brain will be fatal to the OPCs. This needs to be tested with iPSCs before it can be considered for clinical purposes.