Membership Signup
Singularity University

Scientists Convert Human Skin Cells to Neurons For First Time

Marius Wernig and colleagues at Stanford successfully transdifferentiated human skin cells into functional neurons–and they only needed 4 genes to do it.

As if gathering speed toward some uncertain but reachable destination, scientists have checked yet another item off the grand list of things to do before stem therapy is made reality. Marius Wernig’s group at Stanford University became the first to successfully convert mature human skin cells into functional neurons while avoiding the induced pluripotent stem cell stage. The accomplishment gives us reason for optimism, but also warns of the challenges that lie ahead.

The study comes on the heels of an already remarkable milestone recently achieved by the Wernig lab. In January 2010 the group successfully converted mature skin cells they’d gotten from mice to neurons. Incredibly, only three genes were needed for the conversion: Asc11, Brn2, and Mytl1. Affectionately known as “BAM,” these three genes were among a group of candidates they tested known to induce the differentiation of stem cells to neurons. They attached the BAM genes to virus vectors and infected the skin cells, and in a matter of days they had a dish full of neurons. It was the first time skin cells had been converted to neurons in a lab. The reprogramming process followed a protocol that is increasingly being used to convert skin cells to other cell types. Known as transdifferentiation, the procedure skips the heretofore common practice of first inducing the mature cell to stem cell pluripotency and goes straight to differentiation. The reprogramming victory in mice cells set the stage for an attempt with human cells.

Wernig’s group derived skin cells from aborted fetus tissue and the foreskins of newborns. Initially, the BAM combination appeared to have worked. But upon closer inspection, the cells in the dish that looked like neurons were incapable of the electrical communication that is the essence of neuronal function. Back to the drawing board, Wernig’s group pulled another gene off the neuronal differentiation list: another transcription factor called NeuroD. It did the trick. Four to five weeks after infection, they had a dish of neurons that expressed the proper proteins, were electrically active, and formed synapses with other neurons. When plated together the human neurons even formed synapses with mouse neurons. The triumphant conversion of human cells moves the field of regenerative medicine one step closer to patient-specific replacement of lost or dysfunctional neurons. The procedure could also enable scientists to study neurons from patients with neurological disorders such as Parkinson’s disease or Alzheimer’s.

Skipping the iPS stage, as we’ve mentioned before, has multiple advantages. For one, the transcription factors that drive reprogramming can also cause tumors. By deactivating the genes before the–in this case–skin cell is induced to full pluripotency, the risk of tumor generation is decreased. On top of that, the iPS-based procedure is labor-intensive and time consuming. Earlier this year a group at Scripps Research Institute converted mouse skin cells to beating heart cells. Avoiding the iPS stage cut the time from weeks to days. Further darkening the iPS approach, in a recent study mice rejected induced stem cells derived from the skin cells of a genetically identical mouse. It is thought that the rejection was caused by the transcription factors used to induce the iPS cells. Scientists and clinicians were left to wonder if indeed patient-specific iPS cells would one day be a viable treatment.

Not too long ago, this neuron was a skin cell.

There were differences between the two Wernig studies that highlights the higher complexity of human cells to mouse and may portend a challenging road to treatment. As already noted, coaxing the human skin cell into a neuron required an additional fourth gene. Another difference was reprogramming efficiency. Only 2 to 4 percent of the human skin cells became neurons. That’s about 8 percent of the efficiency they’d had while converting mouse cells. More troubling is the fact that almost all of the nascent human neurons communicated with the sole neurotransmitter glutamate. In the brain alone there are over 100 different neurotransmitters that mediate the innumerable subtleties of neuronal function. Scientists will be severely limited in the diseases they are able to study or treat until they find ways to produce neurons that produce the neurotransmitters gone awry in those diseases, such as dopamine in Parkinson’s disease. With these snags in mind, Wernig’s group is currently focusing their efforts on optimizing their protocol to produce more numerous and diverse neurons.


More often than not in science we overestimate progress in the shortterm but under-estimate progress in the longterm. It’s a rule of thumb in the lab that however long you think it will take to finish something–triple that. But stem cell researchers seem to be ignoring the rule and knocking down their objectives as fast as we science writers can describe them. It wasn’t very long ago at all that the first mature cells were successfully induced to pluripotent stem cells. But in the five years since that demonstration was carried out in mice, the iPS field has already been succeeded by the field of transdifferentiation which is now charging forward with a torrent of its own. In just the past year researchers for the first time transdifferentiated skin cells into heart, blood, and liver cells. As my training was in neuroscience, I’m particularly excited about the latest entry to the transdifferentiation club. But the field is and will continue to be exciting to watch regardless, and I can hardly wait to see what they come up with next.

[image credits: Stanford University and Nature]
image 1: Wernig
image 2: Nature


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>


  • Paul Rattner says:

    I can imagine a science fiction assassination:

    So, what would happen if you gave someone a big injection of the gene carrying virus? Why, all their skin cells would transform into pain receptors, causing the most agonizing death possible over a course of weeks! We’ve built a syringe into the end of this umbrella here…

    Worse yet, what if it was attached to a virulent, highly contagious airborne virus, like Influenza? 90% of the world would dissolve into agonized masses of raw nerve goo in weeks!

    Just kidding :) This is indeed miraculous. It’s just, when you read stuff like this, other possibilities seem to suggest themselves…

  • Valdemar Zetterberg says:

    well, a research group in Lund, Sweden, have found a way to convert fibroblasts to dopaminergic neurons. check the latest edition of PNAS (june 6)

Singularity Hub Newsletter