Magdalena Goetz
Helmholtz Center, Germany

This fascinating work demonstrates, for the first time, the full conversion of young fibroblasts into functional neurons by a mix of at least 3 transcription factors.

Compared with previous work demonstrating the full conversion of young astrocytes into functional neurons reaching up to 80% efficiency with a single transcription factor (see {1,2}, on both of which I am an author), the efficiency of conversion from fibroblasts in this paper is much lower (ranging between 2-20%) and requires transduction with several transcription factors. However, the amazing news is that such direct reprogramming even works across germ layers (mesodermal-derived fibroblasts into ectodermal-derived neurons). Interestingly, ectodermal-derived glioma cells typically acquire a mesodermal gene expression signature {3}, pointing to the possibility of a relative 'easy' transition in the transcriptional programs between neuro-ectodermal and mesodermal cells, which may also explain the surprising success in converting fibroblasts into neurons.

References: {1} Heins et al. Nature Neurosci 2002, 5:308-15 [PMID: 11896398]. {2} Berninger et al. J Neurosci 2007, 27:8654-64. PMID: 17687043]. {3} Carro et al. Nature 2010, 463:318-25 [PMID: 20032975].

----

Arnold Kriegstein
University of California, United States of America

This is an interesting paper in which mouse keratinocytes from skin and tail are induced to produce cortical neurons through transfection with three transcription factors. This breaks down yet another barrier to reprogramming one somatic cell type into another and provides a non-stem cell approach to creating patient-specific nerve cells.

Vierbuchen and colleagues tested combinations of lineage-specific transcription factors to directly convert mouse fibroblasts from pre-natal embryos or post-natal tail-tips into functional neurons. They started with a pool of 19 candidate genes and found that a combination of only three genes, Ascl1, Brn2 and either Myt1l or Zic1, was sufficient to convert fibroblasts into nerve cells with an efficiency of around 20%. These 'iN' (induced neuronal) cells expressed a variety of neuronal markers, were capable of firing action potentials and made synaptic connections. The door is clearly open for exploring what other cell types can be generated through the activation of distinct transcription factor combinations.

----

Andy Groves
Baylor College of Medicine, United States of America

This paper shows for the first time that undifferentiated mesodermal cells can be re-programmed to generate functional neurons by the expression of just 3 transcription factors. This work has enormous implications for understanding neuronal differentiation and epigenetic reprogramming.

The authors tested the ability of a cocktail of 19 different transcription factors to convert mouse embryonic fibroblasts into neurons. They ultimately showed that just 3 - Ascl1/Mash1, Brn2 and Myt1 could achieve conversion into neurons capable of forming action potentials and exhibiting aspects of synapse formation. This triple cocktail can also convert postnatal fibroblasts from the tail into neurons. In addition to being an exciting technological advance, this study raises several interesting questions for the future. For example, can ‘re-programmed’ neurons (iN neurons) be induced to adopt particular neurotransmitter phenotypes, by environmental signals, in a dish or after transplantation, or will additional transcription factors be required? What relationship do the iN neurons have to neurons in the central nervous system (CNS), given that factors like Ascl1 are only expressed in subsets of CNS neurons? Finally, is there something especially plastic about undifferentiated fibroblasts that predisposes them to this sort of reprogramming, or will it be possible to reprogram more differentiated cells from different germ layers into iN cells?