Researchers have developed a technique to convert mouse and human fibroblasts directly into dopaminergic neurons without having to first generate induced pluripotent stem cells (iPSCs). They claim a set of just three transcription factors is sufficient to transform prenatal and adult fibroblasts from healthy donors and Parkinson disease patients into functional dopaminergic neurons that release dopamine and show spontaneous electrical activity mirroring that of brain dopaminergic neurons.
Vania Broccoli, Ph.D., and colleagues at the San Raffaele Scientific Institute, describe their approach in Nature. The paper is titled “Direct generation of functional dopaminergic neurons from mouse and human fibroblasts.”
A recent study has demonstrated that fibroblasts can be directly converted into neuronal cells (iNs) by the forced expression of three neurodevelopmental factors, Mash1, Brn2 (also known as Pou3f2), and Myt1l, Dr. Broccoli and colleagues report. However, they note, the resulting iNs represent a heterogeneous population of glutamatergic and GAβAergic neurons, and it’s not yet clear whether specific neuronal subtypes can be preferentially induced by direct reprogramming of these heterologous cells.
The San Raffaele Institute researchers' aim was to generate dopaminergic neurons specifically through the direct conversion of somatic cells by forced expression of lineage-specific factors that act during brain development. The approach was to transduce mouse embryonic fibroblasts from TH-GFP with Mash1, which plays an essential role as a proneural gene during neurogenesis, with each of 11 dopaminergic factors and three iN factors.
The TH-GFP mouse expresses GFP in midbrain dopamine neurons under the control of the rat tyrosine hydroxylase gene promoter and is used as a model for visualizing dopamine neurons. In these initial experiments reporter gene expression was elicited only when Mash1 was combined with Nurr1, a critical determinant of dopaminergic neuronal specification and survival during development and in adulthood. However, this dual combination of Mash1 and Nurr1 was only mnodestly effective in generating GFP+ cells.
The next stage was to combine Mash1 and Nurr1 with each of the other factors in turn. Only the addition of Lmx1a or, to a lesser extent, Lmx1b to the Mash1/Nurr1 mix robustly increased the generation of GFP+ cells with an evidently complex neuronal morphology.
Subsequent addition of a fourth factor failed to generate any further increase in GFP+ cells. In fact, the other two iN factors, Brn2 and Myt1l, reduced the overall reprogramming efficiency. The researchers therefore carried out the rest of their work with cells reprogrammed exclusively with the Mash1/Nurr1/Lmx factor combination.
They found that 16 days after reprogramming, a large number of the resulting GFP+ cells expressed many of the distinctive components of the dopaminergic machinery, such as TH, vesicular monoamine transporter 2 (VMAT2; also known as SLC18A2), dopamine transporter (DAT; also known as SLC6A3), as well as aldehyde dehydrogenase 1a1 (ALDH1A1), and calbindin. Many genes associated with a dopamergic neuronal phenotype were highly enriched.
Conversely, markers associated with adrenergic and serotonergic neurons were not induced in the reprogrammed cells. Also, genes coding for adrenergic and serotonergic biosynthetic enzymes were not upregulated. Fibroblast markers were also downregulated in the iDA cells, while transcriptional analysis by RT–PCR confirmed activation of the dopamine-specific gene network, including the endogenous expression of Nurr1 and Lmx1a.
Moreover, Th and Vmat2 promoter regions were highly demethylated in dopaminergic neuronal cells, whereas they were fully methylated in parental fibroblasts, indicating the epigenetic reactivation during dopaminergic neuronal conversion. The team admits that iDA expression profiling was close to but still distinguishable from that of mDA neurons: 160 genes were differently expressed with a greater than 5-fold change. “Whether this might indicate the presence of residual fibroblast gene expression in iDA cells remains to be addressed,” the researchers note.
Encouragingly, GFP+ cells induced by the three factors showed an elaborate neuronal morphology with multiple and long processes and demonstrated localization of synaptic proteins suggesting the establishment of dopaminergic synaptic terminals. Compared with primary mDA neurons, the GFP+ iDA cells exhibited normal resting membrane potential, normal Na+ currents, overshooting action potentials, and even more prominent K+ currents and afterspike hyperpolarizations. Over 80% of iDA cells showed rhythmic discharges at an average frequency of 2.6 Hz.
Like mDA neurons, the iDA cells in addition expressed high levels of the D2 receptor. Application of a D2/D3 receptor agonist markedly suppressed neuronal firing in 6 of 10 recorded cells in a reversible manner. HPLC measurements confirmed that iDA cells contained high levels of intracellular dopamine, which could be released into the extracellular medium by stimulation with KCL.
To test the potential of the iDA cells in vivo, the team transplanted transduced cells into neonatal mouse brains. Two and six weeks after transplantation, GFP+ cells were found integrated in the host tissue. Most of the GFP+ grafted neuronal cells were positive for TH, AADC, VMAT2, and DAT, indicating the acquirement of a full neuronal dopaminergic cell fate. The team then validated their reprogramming technique using adult human fibroblasts taken from two healthy donors and two patients with genetic forms of Parkinson disease.
The resulting human iDAs were positive for ALDH1A1, TH, AADC, VMAT2, and DAT and demonstrated electrophysiological properties similar to those of the mouse iDA cells. Cell conversion was in addition stable over time, the researchers note. Critically, the cells were also capable of undergoing depolarization.
“Generation of functional dopaminergic neuronal cells by direct reprogramming opens new possibilities for regenerative therapies for Parkinson’s disease and related disorders,” the authors conclude. “The process described here does not pass through proliferative progenitors that also might be tumorigenic. Thus, these procedures might avoid a dangerous drawback of stem cell therapies while providing a sufficient number of functional dopaminergic neurons amenable for autologous cell replacement therapies.”
Dr. Broccoli’s team does note that other studies published within the last few months described the use of a different gene cocktail to induce the formation of dopaminergic-like neurons directly from human fibroblasts. As the San Raffaele team points out, “this opens the intriguing possibility that different molecular fate determinants reach a similar endpoint even though they rely on different transcriptional cascades.”