BERKELEY -- New gene discoveries reported this week are leading researchers to a better understanding of the biological ingredients needed for normal nerve growth, and possibly for the rewiring of severed spinal cord nerves.

In animals ranging from tiny roundworms to humans, timely activation or inactivation of these genes in nerve cells results in the formation of normal nerve connections in the spinal cord or in comparable structures in invertebrates, the new findings suggest.

A collaborating team of Howard Hughes Medical Institute researchers at the University of California at Berkeley and the University of California at San Francisco report the new gene discoveries in two articles in the January 23 issue of Cell and in a third article in the January issue of Neuron.

"Ten years ago we knew very little about molecules guiding nerve-cell growth," said Corey Goodman, professor of neurobiology and genetics at UC Berkeley. "What a difference a decade makes!

"Just within the last five or six years there has been tremendous progress, though we are still far from understanding the complete set of molecules involved in wiring the brain and nervous system and how they work together."

The genes, found in worms, in fruit flies (Drosophila), rats and humans, encode a family of related proteins that mark the extending tips of growing nerve filaments and appear to prevent these nerve filaments from growing into certain areas of the developing embryo.

The researchers' studies of fruit flies and worms indicate that this protein marker is detected by cells that run down the middle of these animals' version of an embryonic spinal cord. Depending on their surroundings, nerve filaments deciding which way to grow may thread their way along adjacent nerve cells, to which they are attracted, or they may follow the less attractive, possibly repellent midline cells. The researchers also found evidence that the protein markers are likewise used in guiding nerve growth around the vertebrate spinal cord.

These new results and other recent findings are leading neurobiologists who study how the nervous system forms to acknowledge the importance not only of growth factors, but also of molecular inhibitors that block nerve growth.

At the same time, an improved understanding of how growing nerves are guided raises hope that the biological processes involved may one day be manipulated to direct the re-growth of nerves to repair severed nerve connections following injury, the researchers say.

Growth factors, growth inhibitors and guidance molecules may all be required to obtain useful re-growth of injured nerves in the spinal cord, according to Marc Tessier-Lavigne, PhD, HHMI investigator and professor of anatomy at UCSF, and a participant in the research collaboration.

"To obtain repair of severed nerve connections in the spinal cord it will be necessary to block the inhibitory molecules -- to inhibit the inhibitors," Tessier-Lavigne says.

The hard-wired scaffolding of the nervous system is laid down in the embryo, and in each organism generates a distinctive nerve-network architecture. In animals with brains, and even in simpler creatures, life experiences add a filigree of new nerve connections to this scaffolding. But many organisms seem to lack the ability to re-form these most basic connections if they are lost due to injury of the brain or spinal cord.

Despite their anatomical differences, organisms share a need to receive and relay sensory information from all sides and to move in response to this information. These movements must be coordinated by motor nerves on both sides of the body. To enact these responses, nerves must be able to relay signals from one side of the body to the other side.

The family of genes now reported by the HHMI research collaborators helps accomplish this wiring task in the embryonic nervous system. The genes are shared by organisms ranging in complexity from worms to humans, suggesting that they arose very early in evolution and have been maintained ever since.

The genes are named roundabout genes (or robo for short) based on the behavior of growing nerve filaments, called axons, in mutant fruit flies discovered about five years ago in the laboratory of Corey Goodman, PhD, an HHMI investigator and professor of neurobiology and genetics at UC Berkeley, and the senior author on two of the newly published studies.

The axons in the mutant fruit flies grow in circles back and forth across the midline of the body, taking the appearance of an English driving circle, or roundabout. The gene was named by an English postdoctoral fellow in the Goodman lab, Guy Tear, who discovered it with another postdoctoral fellow, Mark Seeger.

In newly reported studies conducted in Goodman's lab by Thomas Kidd, PhD, a postdoctoral fellow, along with Tear, who now manages his own laboratory at Imperial College in London, the researchers cloned the roundabout gene in the fruit fly. They discovered that it encodes a new kind of receptor protein, and that the normal version of this protein prevents axons exhibiting it on their surfaces from crossing the midline.

In normal development, axons growing from nerve cells in which the roundabout gene initially is inactive lack the protein marker encoded by the gene, and undetected, they are free to cross the midline. But once they cross, the roundabout gene is activated within the cell and the axons lose the ability to re-cross. The activity or inactivity of the roundabout gene is crucial in the formation of the nervous system, the researchers have demonstrated.

Kidd and Tear also learned that another key gene, commissureless, identified earlier by Tear and Seeger, encodes a protein, Comm, that interacts with the Robo protein and regulates the levels of Robo on the growing axon tips. These two proteins collaborate to control whether axons do or do not cross the midline.

In the worm C. elegans, Jennifer Zallen, a graduate student with the laboratory of HHMI investigator Cori Bargmann, PhD, associate professor of anatomy at UCSF, independently cloned a gene, sax-3, that is similar to robo. Zallen discovered an abnormal version of sax-3 while screening worm populations for mutants that affect axon growth and observing a similar roundabout-like effect. Activation of sax 3 in the worm's axon bundles -- the closest thing to a brain in the worm's anatomy -- is required to guide both motor and sensory nerve axons toward the worm's midline, the researchers report. But once an axon reaches the midline, the presence of the Sax-3 protein marker prevents it from crossing over.

The Robo/Sax-3 protein sits like a TV antenna on the axon surface, recognizing signals from outside the cell and telling the inside of the cell which way it should turn when it reaches the midline.

Katya Brose, a graduate student in Tessier-Lavigne's lab, along with researchers in Goodman's lab, searched for mammalian genes that were similar to the fly robo gene. They discovered human and rat versions of the robo gene, and Brose found that the rat gene was active in nerve cells that approached the midline of the spinal cord. Brose's circumstantial evidence suggests that the Robo protein guides axon growth along the midline in all animals, according to the research collaborators.

The control of nerve growth has long been a mystery, but some general patterns are beginning to emerge. Growth is guided in part by long-range chemical attractants, first proposed to exist a century ago by a giant in the field of neuroanatomy, the Spanish Nobel Laureate Santiago Ramon y Cajal. However, it was just four years ago that long-range chemical attractants were finally isolated and described by Tessier-Lavigne and colleagues.

Tessier-Lavigne's research group discovered two members of a previously unknown class of growth-guiding molecules, named "netrins," from the Sanskrit word meaning "guide." Netrins are needed to form accurate connections between nerve cells that transmit sensory signals through the spinal cord.

The netrins were discovered to be evolutionarily preserved in worms, flies and vertebrates, just as the roundabout genes, which act as short-range repellents, now have been shown to be similarly preserved across phyla.

"We would like to understand not only the molecules involved, but also the logic of how connectivity is encoded," Goodman says.