Are there chemical repellents, as well as attractants, that guide the axons?

   Tessier-Lavigne: It turns out that the netrins are also repellents. That particular insight really came from the nematode. In UNC-6 mutants in nematodes, there is misrouting not just of axons that grow toward the source of the UNC-6 protein, but also of axons that grow away from the source. The simplest way to rationalize that effect is to hypothesize that UNC-6 might also be a repellent. This prompted us to test vertebrates to see whether axons that grow away from the source of netrin are repelled by netrin. We found that this is indeed the case for motor axons that grow away from the floorplate, which is a source of netrin. So the current model is that the netrins and UNC-6 are bifunctional molecules. They can attract some axons and repel others, apparently depending on the nerin receptors made by the axons.

  Do the netrins and UNC-6 account for all the axonal guidance factors discovered?

   Tessier-Lavigne: Definitely not. For instance, there are the semaphorins. One member of that family, called semaphorin III or collapsin-1, is a very potent repellent of a number of different classes of axons. It is hypothesized that different members of that family may actually function quite generally as repellents. And there are many others. There is in vivo evidence, for instance, for an involvement in axon guidance of molecules that are members of the immunoglobulin superfamily and which function as cell adhesion molecules. Similarly, the ephrins, which are ligands for Eph receptor tyrosine kinases, have been implicated in repulsive axon guidance. Yet other molecules have been shown to have potent in vitro effects. Laminin, for example, when coated on a dish, is a very potent promoter of neuron growth. It is also present in vivo in tracks where some axons grow. So it is reasonable to assume that it actually plays a role in directing the growth of these axons.
   In some cases it has been difficult, for technical reasons, to test the roles of particular candidates. For instance, knockouts of some of these molecules are lethal to the early embryo. You can't readily ask what happens later in the nervous system when the molecule is absent, although there are ways around this problem. It addition, it is assumed that there are other molecules out there that we don't yet know of, because there are many guidance events that we can't account for on the basis of known molecules.

  What are the next experiments you have planned?

   Tessier-Lavigne: Right now we have the rough outlines of how the netrins affect commissural axons in vitro and how, in the loss-of-function situation in vivo, the axons get misrouted. Now we want to get an axon's view of that process: we want to know, for example, what is the exact distribution of netrin proteins being encountered; how does it precisely guide the axons; how does it work with other cues to guide the axons? The other side of the equation is the receptor mechanisms. We currently have one component of a receptor involved in netrin-mediated attraction. We'd like to define the receptor complex more fully and to understand better how it works. When netrin binds the receptor, how is it that it stimulates and orients growth? We're also trying to identify other attractants and repellents and their mechanisms of action in the nervous system.

  In your writing, you've described the field as confronting three major challenges and one mystery. What are they?

   Tessier-Lavigne: In a recent review, Corey Goodman and I summarized these as follows. The first challenge is to identify other guidance cues and their receptors; that is, to get a clearer picture of what the major attractants and repellents and their receptors are. We've just barely started to scratch the surface there. The second challenge is to more fully understand what these cues are doing in vivo. We already know there's a certain amount of redundancy—for instance, situations where an attractant is pulling the growth cone from afar and a repellent is pushing it from behind. We need to understand better how these combinations of cues are organized, and what the specific roles of individual molecules are. The third challenge is to deepen our understanding of how growth cones sense these molecules. We have some candidate receptors now but only limited understanding of how the process proceeds frmligand binding to actual growth-cone steering and turning of the growth cone.
   The major mystery is how these guidance cues are used for recognition of specific target cells by the axons. One example is provided by motoneurons, which encounter a field of muscle cells and have to hook up to just the right ones. That's been studied in detail in insects, and there is clearly a very high degree of precision in the selection of targets. One might have expected that each muscle cell would have a particular label that is recognized in a lock-and-key fashion by a receptor on the appropriate motoneuron, but the available evidence suggests that this is not the case. What we know so far suggests that different types of molecules of different varied structures—some of them repellent, some attractant—are arrayed in overlapping patterns that do not make any obvious sense, to direct these very specific decisions. It's almost as if cues are being pulled together at random to jury-rig this very precise target selection that occ. So the logic of how guidance cues are used to direct target selection is, to me, a very big mystery right now. The other things are challenges because we know the outlines of the answers and we just need to do more work.

  It seems that there should be some profound clinical implications for axonal guidance factors.

   Tessier-Lavigne: Most definitely. One area that these studies are likely to impinge on is the normal plasticity of the adult brain. The mechanisms involved in the initial wiring of the nervous system and the selection of target cells are likely to be similar, if not identical, to the mechanisms involved in the normal rewiring of the nervous system that is thought to occur, for example, in learning and memory—when there clearly can be rearrangements of the connections between neurons and their target cells in the adult. This is one area where this work could go beyond an understanding of embryonic development to possibly help elucidate some cognitive functions.
   Another area is nerve regeneration—spinal cord injury, for example. Paralysis results when axons are damaged or severed. To regain motor control it may be necessary for the axons to regrow to their targets in the spinal cord. Molecules that can stimulate growth are likely to be clinically useful in stimulating regrowth of axons in that context. And just as important—a major lesson we've learned in the last 10 years—is that in the adult, it's clear that regeneration does not occur, because of both the absence of stimulators and the presence of molecules that actively inhibit growth. If we could identify those inhibiting molecules and find ways of blocking their action, we might be able to stimulate regeneration as well. So we need to identify stimulators and inhibitors, and that's what's coming out of this field.

Search | November/December 1997 Index | Archives | Contact | Home