Previous studies used several genetic factors and chemicals to perform the same feat but scientists in the USA report that just one gene, Sox2, is sufficient.
Researchers from the Gladstone Institutes and the University of California, San Francisco (UCSF) transferred Sox2 into skin cells taken from both humans and mice. Within days, these cells transformed into early-stage brain cells called induced neural stem cells, which have the potential to grow into any cell type found in the brain.
The cells then began to self-renew and mature, and within a month had formed fully functioning networks of neurons and associated cell types.
In most previous studies where skin cells have been transformed into brain cells, the skin cells were first changed into induced pluripotent stem (iPS) cells. These cells, which are similar to embryonic stem cells, can develop into almost any cell type in the body, and would be induced to turn into brain cells.
The problem with this approach is that 'rogue' iPS cells can form tumours in live animals, meaning that they might be unsuitable for eventual medical use. The researchers hoped they could avoid that risk by turning the skin cells directly into neural stem cells, without passing through a pluripotent state.
'We wanted to see whether these newly generated neurons could result in tumour growth after transplanting them into mouse brains', said Karen Ring, a PhD student at UCSF and the lead author on the paper. 'Instead we saw the reprogrammed cells integrate into the mouse's brain - and not a single tumour developed'.
In the long term, this approach could be used to generate new brain cells to replace those lost in neurodegenerative diseases like Alzheimer's disease or Parkinson's disease. But by allowing scientists to grow functioning human neurons in a lab dish, manipulation of the Sox2 gene may also become important to researchers investigating other aspects of brain function.
The team say that they hope to find further genes allowing them to generate specific types of neurons; for example the dopamine neurons that perish in Parkinson's disease.
'Many drug candidates - especially those developed for neurodegenerative diseases - fail in clinical trials because current models don't accurately predict the drug's effects on the human brain', said study leader Dr Yadong Huang from the Gladstone Institutes.
By generating certain types of neuron, Dr Huang explained, researchers 'could then test drugs that affect different neuron types, [...] helping us to put drug development for neurodegenerative diseases on the fast track'.
The study is reported in the journal Cell Stem Cell.