The slow route to motor neurons has been a “bottleneck,” wrote Vania Broccoli of the San Raffaele Scientific Institute in Milan, Italy, in an e-mail to ARF. “This is a great advance that will likely speed up research on in-vitro modeling of motor neuron diseases.”

The classic method to turn stem cells or induced pluripotent cells into motor neurons involves a 10-day protocol to convert them to neural progenitors, then treatment with retinoic acid and the signaling protein sonic hedgehog to make a motor phenotype (see ARF related news story on Wichterle et al., 2002 and ARF related news story on Dimos et al., 2008). With human cells, this protocol takes six to eight weeks, and “you are lucky to get 30 percent motor neurons,” Kaspar lamented.

The researchers sought a shortcut, looking to transcription factors that are activated to mediate motor neuron differentiation and maturation once the cells respond to retinoic acid and sonic hedgehog. Collaborating with Samuel Pfaff of the Salk Institute in La Jolla, California, they determined that a heterotrimer of transcription factors is key to the motor neuron pathway. Neurogenin 2, islet-1, and LIM/homeobox protein 3 work together to transform neural precursors into full-fledged motor neurons. In particular, they turn on a further transcription factor, HB9, which controls much of the motor neuron transcriptome.

Normally, progenitors make the three transcription factors two or three weeks into the differentiation process. To jump-start differentiation, Hester used an adenovirus vector to deliver genes for the three at the same time as he added sonic hedgehog and retinoic acid. Within just 11 days, the cells contained HB9 as well as choline acetyltransferase (CHAT), another motor neuron marker. Another bonus was that nearly 70 percent of cells went down the motor neuron pathway. By 15 days, the scientists confirmed that the neurons could maintain action potentials, and possessed the strong sodium currents typical of motor neurons.

To further test function, the researchers cultured their homemade motor neurons with muscle cells, looking for the formation of neuromuscular junctions. They were pleased to see that motor axons mingled with the muscle, colocalizing with acetylcholine receptors and producing synaptic markers. Thus, the authors concluded that they had produced mature motor neurons inside of a month.

Using transcription factors to control fate is a good idea, said Zhiping Ping of Stanford University in Palo Alto, California, but as to “whether the cells are mature or not, I am not quite convinced,” he said. He noted that the authors did not provide evidence that the motor neurons' signal is actually received by the muscle cells. Moreover, they have not yet shown that the motor neurons can receive upstream signals, from interneurons, for example.

There are hundreds of different kinds of motor neurons in the spinal cord. Following this recipe, Hester made motor neurons that express homeobox transcription factors matching those in cervical motor neurons, which innervate upper body muscles. Motor neurons made with retinoic acid and sonic hedgehog typically connect with the axial muscles of the trunk and head, Kaspar said. However, with this “proof of concept” in hand, he suggested other transcription factors could make other kinds of neurons. For example, researchers have used the transcription factor Lmx1a to force neurons toward a dopaminergic fate. “A similar approach is conceivable for a fast-track generation of other neuronal subtypes,” Broccoli agreed. In the meantime, Hester can spend more time studying motor neurons and less time making them.—Amber Dance.

Reference:
Hester ME, Murtha MJ, Song S, Rao M, Miranda CJ, Meyer K, Tian J, Boulting G, Schaffer DV, Zhu MX, Pfaff SL, Gage FH, Kaspar BK. Rapid and efficient generation of functional motor neurons from human pluripotent stem cells using gene delivered transcription factor codes. Mol Ther. 2011 Jul 19. Doi: 10.1038/mt.2011.135. Abstract