 |
Molecular Mechanisms of Axon Guidance and Target Recognition
Summary: Corey Goodman is a developmental neurobiologist who uses genetic analysis to try to understand the mechanisms that control the wiring of the brain, that is, how neurons find their correct targets, make appropriate synaptic connections, and adjust and refine the size and strength of those synapses.
Our research is aimed at discovering the molecular mechanisms that control the wiring of the brain during development, that is, how neurons find their correct targets, make appropriate synaptic connections, and adjust and refine the size and strength of those synapses. To use the power of genetics to dissect the molecular mechanisms that control brain wiring, we turned to the fruit fly Drosophila. In so doing, we have discovered some of the key genes, gene families, and signaling mechanisms that control axon guidance and target recognition. The success of our approach is based on the remarkable evolutionary conservation of gene structure and function across species. Our discovery of “wiring” genes in Drosophila has served as a springboard to identify their counterparts in mammals. These genes are likely to control important aspects of human brain development and, in addition to providing insights into brain wiring, may hold the key for clinical insights into neurological disease and the repair of nervous system injury.
Midline Guidance, the Robo Code, and Positional Information
With a concerted effort in several laboratories and a variety of organisms, midline guidance has emerged as the best-understood complex guidance decision in any organism. We know many of the key ligands and receptors, some of the regulators, and some of the signal transduction components. The midline, a special group of distinctive cells, separates the two symmetric halves of the developing central nervous system (CNS). Most CNS growth cones confront the midline and either grow toward or away from it. Once near the midline, they decide to cross or not to cross the midline. If attracted to enter the midline as an intermediate target, they do not stay there, but cross it and exit on the other side. Those that do cross the midline, never cross it again. After crossing the midline, they turn in a tract at a stereotyped distance from the midline.
Starting more than a decade ago, we took a genetic approach to this problem in Drosophila. Around the same time, Marc Tessier-Lavigne (HHMI, University of California, San Francisco), starting when he was a postdoctoral fellow with Tom Jessell (HHMI, Columbia University), took a biochemical approach to the same problem in vertebrates. Some of our studies are collaborations between the two laboratories, as together we move back and forth between species.
Midline cells in Drosophila secrete two guidance cues: Netrins and Slit. Growth cones can be either attracted toward or repelled away from these two signals. The combination of receptors on a growth cone’s surface endows it with its distinctive responsiveness to these cues. For example, DCC (called Frazzled in Drosophila) is an attractive Netrin receptor. Growth cones expressing DCC are attracted toward the midline by Netrins. Roundabout (Robo) is a repulsive Slit receptor. Axons that never cross the midline express high levels of Robo from the outset. Growth cones that do cross the midline dramatically increase their levels of Robo after they cross. Commissureless (Comm) negatively regulates the levels of Robo by driving internalization of the receptor. Comm2 functions with Comm to regulate the levels of Robo.
Attraction versus repulsion is encoded in the cytoplasmic domains of these receptors. For example, DCC-Robo, a chimeric receptor containing DCC’s ectodomain and Robo’s cytoplasmic domain, when reintroduced back into the developing organism, functions as a Netrin receptor that mediates repulsion away from Netrin. The Robo cytoplasmic domain mediates repulsion in part by binding to the Enabled adapter protein. The Abl tyrosine kinase negatively regulates this output of Robo.
When Slit is eliminated, axons are attracted to the midline but do not leave it. However, in the absence of Robo, growth cones freely cross and recross the midline but they do not stay at the midline. The Drosophila genome encodes three Robo family members. When both Robo and Robo2 are eliminated, axons are attracted to the midline, but do not leave it. Thus these two Robos together account for most if not all of the repulsive activity of Slit at the midline.
While Robo and Robo2 control midline guidance, Robo2 and Robo3 play another key role. Some Slit diffuses away from the midline, generating a gradient from the CNS midline to its lateral edge. The Robo code (i.e., the combination of Robo receptors expressed by any particular growth cone) controls the general lateral position at which a growth cone turns to extend longitudinally. Axons that extend medially (closest to the midline) express only Robo. Axons that extend at intermediate locations express Robo2 and Robo, while axons that extend laterally (furthest from the midline) express Robo3 in addition to the other two Robos. Genetic analysis confirms the role of the Robo code in specifying position within the developing CNS. For example, overexpressing Robo2 by axons that normally travel near to the midline causes them to move further away and join lateral bundles. However, the Robo code is not sufficient for precise topography. Rather, our genetic analysis shows that precise topography of longitudinal pathways is controlled by a combination of long-range guidance (the Robo code determining region) and short-range guidance (discrete local cues such as Fasciclin II determining specific location within a region).
We have also discovered that Slit can function as an attractant. Migrating mesodermal cells respond to Slit as both an attractant and a repellent. Interestingly, Robo receptors are required for both functions. Mesoderm cells initially migrate away from Slit at the midline. A few hours later, these same cells change their behavior and extend toward specific attachment sites. During this next phase, Slit secreted by specific muscle attachment sites functions as an attractant for mesoderm cells expressing Robo receptors. Thus Slit is bifunctional. Moreover, over a period of a few hours, individual cells in the developing organism switch their responsiveness to Slit from repulsion to attraction. (Grants from the National Institutes of Health and the American Paralysis Association provided partial support for these projects.)
Semaphorins and Their Plexin Receptors
Some years ago, we discovered the Semaphorins, the largest family of repellents in humans. Although they were originally described as repellents, we and others have found that numerous Semaphorins can also function as attractants. In collaboration with Tessier-Lavigne’s lab, Melanie Spriggs (Immunex), and Paolo Comoglio’s lab (Institute for Cancer Research, Turin, Italy), we found that Plexins are receptors for Semaphorins. We have begun to elucidate the function of Plexins, their role as repulsive receptors, and their downstream signaling. We have found that Plexins mediate repulsive guidance in part by binding to and blocking the output of the active form of Rac, a small GTPase whose output normally increases cell motility.
Motor Axons, Choice Points, and Target Recognition
One particularly accessible target-derived choice point is the ability of motor axons to defasciculate selectively (i.e., de-adhere from other axons) from the common motor pathway and to steer into their appropriate muscle target region. Genetic screens have led to the identification of numerous genes that control these events. Correct entry into the target region requires the expression of a target-derived attractant called Sidestep (Side), a transmembrane protein on the surface of target muscles. Defasciculation also requires Beaten Path (Beat), a protein secreted by the motor axons. Guidance at this choice point is controlled by the balance of attraction and repulsion. The fasciculating motor axons express the cell adhesion molecule Fasciclin II as their major attractant and transmembrane Semaphorins and Plexins as repellents. Contact with the target-derived attractant Side appears to trigger the release of Beat and a shift toward repulsion leading to defasciculation.
Last updated April 10, 2001
|
|
|
|
