Genetics: Telomere Loss Spells Trouble for Aging Mice Cell Biology: Nerve Cells on the Go Neurobiology: How the Nose Knows a Multitude Of Different Odors Ambulatory Care: New Medicare Drug Coverage Proposed for Low-Income Beneficiaries Structure Solved for Cell Death Protein Low Birthweight Linked to Type II Diabetes Mouse Model of Cushing's Disease Developed Species-Jumping DNA May Aid Bacteria Studies Pulse Pressure Predicts Congestive Heart Failure in Elderly Faculty Council: Talk Ranges from Health Services to Authorship In Memoriam: William Alonso Honors & Advances News Briefs Why Tom Wolfe (Almost) Made Me Fail My Prelim Exam Front Page

CELL BIOLOGY Nerve Cells on the Go Axon Pathfinding Tied to Cytoskeleton Research One of the hottest fields in cell biology aims to understand the molecules that drive the cytoskeleton, the gel-like inner scaffold that allows a cell to "morph" into different shapes as it responds to important changes in its environment. And one of the hottest fields in neurobiology aims to understand how the hand-shaped end of a growing neuron, called the growth cone, explores the territory it traverses on its way to its target tissue. For comic relief from their work on the intricacies of the cytoskeleton, David Van Vactor's group has festooned the lab with an oddly fitting mixture of brain, fly, alien, and Elvis paraphernalia, including the hand-blown glass brain above. The twain now meet in two NIH-funded studies published in the February Neuron. In them, David Van Vactor, HMS assistant professor of cell biology, describes how his team found an uninterrupted chain of signaling events that neurons in fly embryos use to transmit outside information from the membrane all the way to the actin cytoskeleton. In previous research, cytoskeleton researchers have worked from the bottom up, tracing their way backward from actin. They have found a bewildering number of partial connections but have not yet made the leap to the membrane receptor. Working from the top down, neuron guidance researchers have traditionally studied which external cues the growth cone encounters and which receptors it uses to recognize them. But this research has not yet completed the link to actin. Van Vactor's work bridges that gap by describing the first continuous line of communication linking a receptor in a growth cone's membrane to actin, the final agent of change in cell shape. The cell's cytoskeleton is a continuously changing fabric of protein filaments underneath the cell membrane (see Focus 9/19/97). Actin is one of its major components. Beyond giving a cell mechanical strength, the cytoskeleton is the executor of most biological events that require a change in a cell's shape or motility. Examples include the development of an organism with different types of cells arranged in the proper places, or the rogue travels of metastatic cancer cells. Lost Nerve, Found Insight Van Vactor started this research as a postdoctoral fellow in the lab of Corey Goodman at the University of California, Berkeley, who is a co-author on the first paper. Using the fruit fly, Van Vactor analyzed two types of mutant phenotypes with derailed motor nerve development. In one set of mutants--dubbed stop short--a motor nerve called intersegmental nerve b (ISNb) arrested its growth before reaching its target muscles, suggesting the disrupted genes were essential for the growth cones to proceed. In mutants dubbed bypass, ISNb neurons miss their exit, growing straight past the muscle instead of turning sharply toward it (see figures b and c). When he cloned the genes underlying stop short and bypass, Van Vactor expected they would operate in different contexts. But once he analyzed them a puzzle fell into place, and the genes turned out to belong to the same pathway. The story starts at the bottom of the pathway. The first gene he cloned caused the stop short phenotype. It turned out to encode profilin, a much-studied protein known to bind and control actin. That made sense and was not really surprising, Van Vactor says. Unexpected, however, was his finding that the stop short phenotype also arose in embryos lacking the gene for the protein kinase Abl. (Protein kinases are enzymes that tack phosphate groups onto other kinds of proteins.) When analyzing mutant embryos that lacked both profilin genes and one copy of the Abl gene, the scientists found that cutting the amount of Abl protein in half dramatically worsened the embryos' stunted nerve growth. This genetic way of asking whether one protein is sensitive to the dose of another helps scientists find out whether two proteins cooperate in the same pathway. Profilin and Abl clearly seemed to do so. Abl provided an intriguing step up the pathway, since its substrate--a protein called Ena--was known from other systems to bind profilin and affect actin. "So it was actin-profilin-Ena-Abl from the bottom up," Van Vactor says. Research in David Van Vactor's lab has unraveled a molecular chain reaction that allows the tip of a growing neuron to turn in the right direction. Motor nerves in a developing fly embryo fan out from its ventral nerve cord toward its back (top) in search of their target muscles (a). In wild type embryos (b), the ISNb nerve veers sharply to the right (arrow) as it reaches its target muscles, but in mutants missing the Dlar protein (c), it shoots straight past. A diagram (d) conceptualizes what happens in the normal growth cone as it encounters an attractive cue (black dot), such as those that might be found on the target muscle. The semicircle above the growth cone serves to plot actin assembly across the growth cone's leading edge (red). Though the simultaneous making and breaking of the actin cytoskeleton continues along the entire front, a local spike in actin polymerization just underneath the attractive cue can gradually push the growth cone in its direction. Two reports in Neuron describe a signal transduction pathway that transmits this behavior internally (e). It includes a phosphorylation-dependent "switch" made of the receptor phosphatase Dlar, the kinase Abl, and the phosphate recipient Ena, as well as the actin-regulator profilin, among others. The second paper exposes the other half of the pathway from the membrane downward. It begins with Dlar, a gene causing the bypass phenotype. Dlar is a member of the receptor tyrosine phosphatase family--membrane-spanning proteins that slice phosphate groups off other proteins inside the cell. Three years ago, Van Vactor and Haruo Saito, HMS professor of biological chemistry and molecular pharmacology, first implicated Dlar in axon guidance. Trying to understand how Dlar signals, Van Vactor, working with graduate student Zachary Wills and others, discovered that a triumvirate of proteins--Dlar, Abl, and Ena--is bound together in intimate, antagonistic relationships. The trio makes key decisions about what information is transmitted to the cytoskeleton, Van Vactor says. Evidence supporting that idea also comes from experiments testing the sensitivity of one protein to the dose of another. The scientists found that Dlar is as sensitive to the amount of Abl as is profilin. Halving the amount of Abl protein suppressed the damage wrought by the Dlar mutation--that is, fewer ISNb nerves bypassed their targets. Conversely, increasing the amount of Abl protein beyond normal levels overwhelmed Dlar and produced the bypass phenotype in normal fly embryos, just as if they had a Dlar mutation. This key experiment shows that the kinase Abl and the phosphatase Dlar are opposing enzymes locked in a balance of power, and each one can tip the scale. Bringing the research full circle, the scientists show that an object of this competition was the phosphate recipient Ena, the protein known to interact with profilin. A Method to the Madness One reason this work appears complicated is that even though the researchers have established a sequence of players, they still do not fully understand the precise relationships among them. It is not as if a valve opened at the membrane, passing information all the way to actin like water flowing smoothly down a tube. On the contrary, there seems to be plenty of turbulence, with members of the pathway frequently opposing their binding partners, as if they were haggling over the final outcome every step of the way. Take these examples: Ena suppresses Abl, Abl fights Dlar, Ena and Dlar work together, but Ena seems to inhibit profilin, and profilin curbs the polymerization of actin. "Even so, a coherent picture is finally beginning to drop out of all this madness," Van Vactor says. Indeed, these kinds of checks and balances, combined with input from multiple tributaries to this flow that have yet to be established, may represent the molecular basis of a "thinking" signal transduction system that integrates multifaceted, and sometimes conflicting, information to come up with the appropriate biological response. Many questions remain unanswered, Van Vactor says, but it seems clear that these interlocking relationships are well suited to make the cytoskeleton as dynamic as it is. The adding and removing of small phosphate groups to signal transduction proteins in the growth cone probably serves as a fast and reversible mechanism, enabling the growth cone to weigh attractive versus repellent cues as it travels the ever-changing landscape of the developing embryo. --Gabrielle Strobel Back to Top