Thursday 7 November 2002
SFN 2002 - Day 5
Investigator: Jane Johnson
by Rabiya Tuma
Conceptually, researchers have known since the 1980s that transcription factors control cell identity. Now, they are learning the exact regulatory pathways that determine why one cell takes on the fate of a particular neuron in the dorsal region of the developing spinal cord rather than that of a neighboring cell.
In the case of the developing dorsal spinal cord, Jane Johnson and colleagues at the University of Texas Southwestern Medical Center in Dallas find that the basic helix-loop-helix (bHLH) transciption factors play a critical role in determining which neuron a progenitor cell gives rise to.
The progenitor cells of dorsal interneurons type 1 (di1) express the bHLH gene math1, while those of di2 and di3 cells express NGN1/2 and mash1, respectively. But how do these different transcription factors change cell fate?
The group has shown that in mice lacking math1, no di1 neurons form. Similarly di2 neurons fail to form in mice lacking NGN1/2. Thus the bHLH factors are necessary for cell -ate identity. Furthermore, overexpression of math1 leads to an excess of di1 neurons at the expense of di2 and di3 neurons, suggesting that it is also sufficient to change the fate of developing neurons.
To test which part of the math1 protein is required for these functions, the group created chimeric proteins switching the helix portions of the DNA binding domain from math1 and replacing it with those of myoD, a bHLH protein involved in muscle cell specification.
When the group replaced helix1 of the math1 protein with the myoD helix1, and electroporated the transgene into the spinal cord of a developing chick embryo, the progenitor cells no longer made neurons, indicating that it is helix1 that specifies neuronal fate. Interestingly, when they replaced the math1 helix2 domain with that of mash1, the progenitor cells made neurons but took on a di3 phenotype, which is the neuronal type specified normally by the mash1 protein. These data indicate that helix2 specifies the neuronal subtype formed.
Additionally Johnson's team is studying the upstream regulation of math1 expression, and have found that it is autoregulated in a positive feedback loop, such that math1 expression induces more math1 expression. They has also identified an inhibitor of math1 called zic1. When excess amounts of zic1 are electroporated into a developing chick spinal cord, it overpowers math1 autoregulation and shuts down math1 expression, leading to a decrease in the number of di1 neurons.
Johnson says this sort of information regarding the regulation of neuronal differentiation is only the beginning. "The autoregulation and zic alone do not explain why we get math1 expressed when and where we do, so there is a lot there we haven't found yet," Johnson told BioMedNet News.
The general principles described in Johnson's work are familiar, says
Eric Turner, a neurobiologist from the University of California in San
Diego. "But at the molecular level no one knows how this works really,"
said Turner. "Here is a molecule that lives in the nucleus and interacts
with proteins and literally turns other genes on or off via a combinatorial
code to make specific neurons. But right now it is almost a black box how
that happens. We just have to get one clue at a time to understand the
molecular pathway and zic1is one clue."
© Elsevier Science Limited 2002