The resulting cells show all the hallmarks of neurons–they express neuron-specific genes, they have the characteristic branching shape of neurons, and they can form electrically active connections both with each other and with regular neurons collected from the brain. “Many people thought it would be impossible to transform cells in this way,” says Zhou. “The fact that you can convert them so rapidly and efficiently is quite surprising.”
Wernig’s team is now trying to replicate this phenomenon in human cells. “If we can accomplish that, it opens the door to entire uncharted areas,” he says. “Then we can derive neurons from a patient’s skin cell, which bypasses the complicated iPS cell process.” IPS cells can be tricky to grow, and the process takes four to six weeks, he says.
It remains to be seen which approach will work best in different situations. One advantage of iPS cells is that they can produce more of themselves, and can therefore be grown indefinitely and in large quantities, says Sheng Ding, a biologist at the Scripps Research Institute, in La Jolla, CA, who was not involved in the current research.
It’s also not yet clear exactly how the remarkable transformation revealed in the latest work happens. Genetically identical cells can have very different identities thanks to epigenetics, which refers to different mechanisms a cell has for packaging its DNA. That packaging regulates which genes are easily accessible and active in the cell, which in turn determines whether it becomes a skin cell, a heart cell, or a brain cell.
Broadly, scientists think that the transcription factors used in various reprogramming recipes alter this DNA packaging. “We need a real epigenetic and molecular understanding of the mechanism in order to manipulate the system more intelligently,” says Zhou.
The mechanisms underlying direct reprogramming may prove more complex than in iPS cell reprogramming. Converting adult cells to an embryonic state may simply involve stripping epigenetic markers. “But when directly reprogramming from one somatic cell to another, you cannot randomly remove epigenetic marks,” says Zhou. “You have to remove some and add some and keep many intact. Recognizing which to leave alone and which to change is the key.”
Before the technology can be tested for human therapies, researchers will likely need to find a combination of chemicals that can achieve the same results as the genes used in the study, because genetically engineered cells may harbor some cancer risk (scientists have already accomplished this with iPS cells). Researchers will also need to show that the cells can function properly when transplanted into the brain–Wernig now plans to test this in mice engineered to have a disease similar to Parkinson’s.
The research is also likely to provoke some rethinking of cell fate. “For a long time, epigenetic modifications were thought to be extremely stable,” says Wernig. “Before Dolly the sheep or iPS cells, people thought epigenetic modifications were irreversible–that once set during development, they were not changeable. But this is absolutely not true.”