It’s difficult for people to change their identities or careers, but it can be done. We don’t have to be stuck with one particular fate; with a bit of effort, we can become different people. The same is true for the cells that we’re made of. They come in different types, from brain cells to skin cells to muscle cells. Stem cells can produce all of these types, but once a cell commit to a particular role, it’s largely stuck there.
But not always. Scientists can convert one type of cell into another with the right cocktail of molecules – a process known as transdifferentiation. It’s a cellular makeover. The hope is that this technique will allow doctors to grow bespoke tissues and organs. If someone suffers from a disease that destroys their nervous system, like Alzheimer’s, you could theoretically take their skin cells, and transform them into a fresh supply of genetically identical neurons.
To do this, you need to work out the right recipe. Many groups are working on this. They’ve managed to change pancreatic cells into liver cells, skin cells into heart cells, and more. But no one has been able to transform other types of cells into human neurons. That is, until now. Zhiping Pang, Nan Yang and Thomas Vierbuchen from Stanford University have identified a quartet of proteins that can change human skin cells into working neurons.
The trio did the same thing with mice last year, using three proteins affectionately known as “BAM factors” – Brn2, Ascl1 and Myt1l. Having proved that this was possible, they turned to human cells. They used viruses to smuggle their protein trinity into skin cells, taken from discarded foreskins. After just over a week, they found that some of these cells had started to look like neurons. They had the right shape and they activated the right genes. But they fell at the last and most important hurdle – they couldn’t carry electric signals in the same way that proper neurons can.
On their own, BAM lacked enough punch. It took a fourth protein – NeuroD1 – to complete the transformation. With their quartet (BAMN) Pang, Yang and Vierbuchen finally managed to fully convert thier skin. Within a few weeks, the born-again neurons were carrying electric currents. They had formed working synapses (the junctions between neighbouring neurons) and were sending signals to one another. And best of all, they could integrate into existing networks. When the team cultured the changing cells alongside existing neurons, they slotted seamlessly in.
Other scientists have also managed to transform skin cells into neurons, but never as directly as Pang, Yang and Vierbuchen have. Instead, they first reprogrammed the skin cells into a stem cell-like state before coaxing them into becoming neurons.
This field has progressed at breakneck speed. In 2008, Rudolf Jaenisch managed to treat Parkinson’s disease in a rat by reprogramming its skin cells, turning them into neurons, and transplanting them back into the rat’s brain. The new neurons integrated into the brains of their hosts and the rats’ symptoms improved. A few months later, Kevin Eggan created neurons by reprogramming skin cells from an 82-year old woman with amyotrophic lateral sclerosis (ALS), the same condition that paralyses Stephen Hawking.
But these cells have a long way to go before they can be used as medical treatments. A wave of recent studies have highlighted risks with the approach. The reprogrammed cells (known as iPSCs) are subtly different from the originals, both in the seqeunces of their genes, and in how those genes are used. Just this month, Yang Xu showed that mouse iPSCs can be rejected by the immune system even though they’re supposed to be genetically identical to their new hosts.
For now, it’s not clear if converting skin cells directly into neurons, without having to go through a “stemmy” intermediate, would work any better. Marius Wernig, who led the study, says that his technique uses different proteins but he suspects that he might eventually run into the same problems with immune rejection and genetic differences.
“However, there is one advantage of the direct approach,” he says. “With iPSCs, you usually make only few lines per patient (typically not more than 3, if that). If that line is not fully reprogrammed, you are screwed.” With his method, you can convert many skin cells simultaneously into neurons. “Some of them will be reprogrammed than others but they are all there to analyze, giving you the entire spectrum.”
Jonathan Slack, who has worked on transforming liver cells into pancreatic ones, says that the main question is whether these cells are really neurons. “Probably yes, as [the researchers] have done a fair bit of physiological analysis.” However, he adds that the BAMN proteins switch on many genes that turn cells into neurons. As a result, the cells might look the part without having actually made a “true developmental switch”.
For Wernig, the next step is to make his transformation more efficient. His team has shown that you can give skin cells a neuronal makeover, but they can’t do it very efficiently and the process takes five to six weeks. Once they’ve sped things up, they want to see if they can create neurons from actual patients with brain diseases.
Slack says that there is a long way to go before we can make human neurons that are fit for transplantation. However, he is optimistic. “Direct reprogramming does offer the prospect of making patient-specific grafts that would not provoke an immune response.”
Slack. sees two main obstacles.Firstly, Wernig’s group (like many in this field) using viruses that shove genes into a host genome. That’s an efficient way of getting the proteins you want into a cell, but it can also increase the risk of cancer. “Then, there is also the issue of controlling the type of neurons that result, as different types will be needed for different purposes. As always there is a long way to go.”
Reference: Pang, Yang, Vierbuchen, Ostermeier, Fuentes, Yang, Citri, Sebastiano, Marro, Sudhof & Wernig. 2011. Induction of human neuronal cells by defined transcription factors. Nature http://dx.doi.org/10.1038/nature10202
More on tailor-made neurons: