November 12, 2004
The State of HSPH:
Gene Expression Profile Predicts Survival in Ovarian Cancer
Birth of Motor Neurons Connected to Spinal Cord Induced in Adult Brain
Five from HMS and HSPH Appointed to IOM
Grants Advance Research on Childhood Brain Tumors
Talking to the Public: How Can Media Coverage of Medicine Be Improved?
Neurons Use Noodle When MotoringBinding Protein Ndel1 a Needed Link in Nerve Cell Migration
Neurons migrate by the millions from the proliferative core of the developing brain to the outer cortex, where they crowd together to create the folds and ridges that give the mature brain its characteristic cauliflower appearance. If the neurons find their travel plans thwarted as happens in the genetic disease lissencephaly, the result is a small, smooth brain, epilepsy, and severe mental retardation. The product of the lissencephaly gene, LIS1, is known to turn on the cell motor dynein, but just how dynein powers nerve cell migration has been a mystery.
Neurons need Ndel1 to power up dynein motors that drive cell migration in brain development. Left to right below, Li-Huei Tsai, Tianzhi Shu, Minh-Dang Nguyen, and colleagues traced the travels of new neurons after RNAi knockdown of Ndel1, LIS1, and dynein expression in fetal mouse brains. A normal neuron (above, top) ties the forward advance of a membrane edge to the transport of the nucleus via microtubule tethers (red lines) that emanate from the centrosome. Their work shows that Ndel1, LIS1, and dynein (triangles and circles) are all required to maintain the microtubule grip on the nucleus. In neurons lacking Ndel1, dynein, or LIS1 (above, bottom), a dropped connection between the centrosome and the nucleus puts the brakes on cell migration. (Photo by Liza Green, HMS Media Services)
HMS researchers used a novel gene-knockdown technique in brains of fetal mice to uncover the critical link between LIS1, dynein, and neuronal migration. Their work shows that the LIS1-binding protein Ndel1 (pronounced "noodle one") supplies the glue that holds LIS1 and dynein together. The trio is required to organize a microtubule network around the nucleus that provides the pull to get neurons where they need to go.
"Ndel1 itself, just like LIS1, is necessary for correct positioning of neurons during cortical development," said Li-Huei Tsai, a Howard Hughes investigator and HMS professor of pathology, who is the principal investigator of the study, which appears in the Oct. 15 Neuron. "If we knock down Ndel1 expression in the embryo, then those neurons cannot end up in the right place." While catastrophic errors in neuronal positioning during development lead to devastating conditions like lissencephaly, subtler defects could also play a role in diseases like schizophrenia, depression, autism, and adult epilepsy.
Snail's PaceMigrating neurons look something like snails, explains Tianzhi Shu, a postdoctoral fellow in Tsai's lab and first author on the paper. A headlike membrane projection leads the way (see figure above), dragging the main cell mass and the nucleus behind. The head and body are linked by two webs of ropelike microtubules, one tied to the membrane at the leading edge and the other coming from the nucleus. The webs join at the centrosome, an organelle that controls microtubule formation.
Experiments using cultured neurons revealed that cells lacking any of the three proteins start to migrate fine, but soon run into trouble. Time-lapse movies show the neurons sending out a normal-looking leading edge headed in the right direction. The advancing front is connected to the centrosome by an apparently normal microtubule network. But the cells fail to form the microtubule handle that links the centrosome to the nucleus, so the nucleus lags behind while the centrosome gets pulled forward and the migration halts.
To study neuronal migration in the developing brain, the researchers also turned to RNAi. They knew that traditional targeted gene knockouts of Ndel1, LIS1, or dynein killed mice before their brains began to develop. To deliver the gene knockout to the exact window of time and space where migration was occurring, the researchers introduced RNAi vectors into mouse fetal brains in utero by electroporation and showed they could acutely decrease levels of Ndel1, LIS1, and dynein proteins over several days. When they looked at the positions of modified neurons, they found that decreasing any of the three proteins resulted in fewer neurons finding their proper homes. Neurons that failed to migrate showed the same abnormal separation of centrosome and nucleus as in the cultured cells.
In utero gene delivery allowed the researchers to manipulate more than one gene at a time and provided evidence that Ndel1, LIS1, and dynein work together in a single pathway to regulate neuron movement. "We did some fancy experiments trying to recapitulate genetic kinds of experiments where we knock down Ndel1 with RNAi and simultaneously overexpress the LIS1 gene. We saw a partial rescue effect that suggests they operate in the same pathway," explained Tsai. These experiments would be hard to do with knockout mice, she added.
LIS1 Double WhammyNdel1 has a sister gene that also binds to LIS1, and recent evidence that it functions in cell division only heightens the interest in the LIS1-centrosome connection in brain development. Mice lacking this LIS1- associated protein, Nde1, have small brains and fewer neurons. "Before this, we thought of LIS1 lissencephaly as just being a problem of migration. This finding raises the possibility that kids with LIS1 mutations also have problems with cell division," said Christopher A. Walsh, a Howard Hughes investigator and the Bullard professor of neurology at Beth Israel Deaconess, who headed the study, published in the same issue of Neuron.
"We're probably looking at two sides of the mirror," Tsai explained. The difference in function between the two Nde relatives certainly reflects their expression patterns: Nde1 is present in dividing neurons, while Ndel1 appears when neurons are ready to migrate. In their own time, each partners with LIS1 and dynein to participate in the centrosome-controlled processes of mitotic spindle orientation or cell migration.
Tsai's next challenge is to learn more about centrosome structure and regulation of the nuclear microtubule network. To do this, she and Shu are working to purify centrosomes to see what other proteins of interest might be lurking there.