Center News

Setting boundaries in the brain

Moens lab finds mechanism that repels cells may also serve to attract, depending upon the context

June 16, 2005
Drs. Cecilia Moens (left) and Hilary Kemp

Drs. Cecilia Moens (left) and Hilary Kemp found a mechanism that is required for normal boundaries to form in the vertebrate hindbrain. The microscope in the photo generated the zebrafish image below.

Photo by Todd McNaught

Many times during animal development, identical cells embark on different pathways, ultimately becoming different types of tissue. To keep cells with distinct destinies away from each other, physical boundaries form between different cell populations.

A recent discovery from Dr. Cecilia Moens' lab in the Basic Science Division reveals a mechanism that is required for normal boundaries to form in the vertebrate hindbrain, which includes the cerebellum and the brainstem. The researchers have shown that a molecular interaction, once thought only to separate different types of cells, can also keep similar cells clustered together. The boundaries that result from this clustering are crucial for the right neurons to develop in the right places.

"Boundary formation is a topic that spans a lot of ground," said Dr. Hilary Kemp, a postdoc in Moens' lab and one of the paper's authors. "We're using these regions of the brain as a model for all kinds of compartmentalization that are important for developmental processes."

The paper — whose first author is Dr. Julie Cooke, a former Moens postdoc who is now at Vastox, Inc. in Oxford, England — was published in the March issue of Current Biology.

Eph-ephrin signaling pathway

One well-established mechanism of sorting cells and establishing boundaries involves proteins called Eph receptor tyrosine kinases. These receptors are embedded in cell membranes and can be mobilized into action by other cell-surface molecules called ephrins.

The Eph-ephrin signaling pathway is involved in many developmental mechanisms, including nerve-cell establishment, segmentation in the brain, and the development of arteries, veins and the lymphatic system.

The Eph receptor was first discovered in cancer cells — "Eph" stands for Erythropoietin-producing hepatoma — but not much has since been learned about how Ephs and ephrins operate in cancer.

"Although these molecules originally were implicated in cancer, their function in cancer isn't particularly well understood, whereas their function in development is becoming much more clear," Kemp said.

Sorting cells into rhombomeres

One place that Ephs and ephrins are known to operate during development is in the vertebrate hindbrain. Early in nervous-system development, different types of hindbrain cells sort from each other into separate brain segments. The result is seven or eight distinct bulges called rhombomeres, which have noticeable boundaries between them. Nerve cells, or neurons, in each rhombomere eventually form nerves and project to different parts of the body.

Researchers have long wondered how hindbrain cells "know" which rhombomeres they belong in. One clue was that Eph receptors and ephrin molecules are often found in alternating rhombomere segments. Building on this observation, many studies have shown that these two molecules repel each other: When a cell with an ephrin on its surface contacts a cell with an Eph receptor, the two cells retract from each other.

Kemp and her colleagues wanted to see if they could find evidence for this repulsion in zebrafish embryos. By injecting specially designed synthetic molecules into the zebrafish, they compromised, or "knocked down," the function of either the receptor or the ephrin and examined the effects in the embryos.

When the Eph receptor was knocked down in normal zebrafish embryos, the boundaries between rhombomeres became noticeably disrupted. Disturbing the ephrin by itself had very little effect, but knocking down both the receptor and the ephrin had a severe consequence: Cellular markers known to be specific to the boundaries between rhombomeres were either in the wrong place or missing entirely.

Outside the standard model

The traditional model of Eph-ephrin signaling would say that the boundaries became disorganized because the Eph and ephrin were not around to repel each other, according to Kemp. To determine whether this was the only mechanism at work, she and her coworkers looked at how cells with no Eph receptor mixed with cells from normal embryos.

If repulsion were the only active mechanism, cells lacking the Eph receptor should drift aimlessly throughout the hindbrain's segments, incapable of being repelled and so mixing evenly with all other cells. Instead, the researchers found that cells missing Eph were excluded from the middle of rhombomeres that normally contain Eph receptors. They moved to edges of segments, near the boundaries, while normal cells clustered tightly together in the middle of these rhombomeres. The cells appeared to need the Eph receptor in order to remain inside their appropriate segment, which hints that those lacking Eph are less able to adhere to other cells.

Compared to cells without their Eph receptor, normal cells "essentially look like they're stickier," Kemp said.

Cooke, Kemp, and Moens propose that this Eph receptor and ephrin repel each other at the boundaries of adjacent rhombomeres, pushing Eph-expressing cells into certain segments and ephrin-expressing cells into others. But, within the rhombomeres themselves, the researchers say, contact among Eph-expressing cells leads to adhesion.

The researchers can't be entirely sure that adhesion and repulsion are equally important for boundary formation, since their experiments eliminated both aspects of Eph-receptor function, Kemp said. But they do know that both mechanisms operate in the hindbrain, and "it's very likely that both functions are important for boundary formation," Kemp said.

Researchers' theories

It is not entirely surprising that cells expressing Eph can adhere to other cells, Kemp said, because recent work has shown that Ephs and ephrins sometimes stick to each other. But she and her colleagues didn't think adhesion was at work in the hindbrain, because no one has found an ephrin expressed in the rhombomeres that contain the Eph-expressing cells. With no ephrins around, the researchers don't know what the Eph receptors could be sticking to.

One possibility for this mystery, Kemp said, is that there is actually an ephrin in the same segments as the receptors, but it's just at such a low concentration that no one has detected it yet. This would fit with some other researchers' theories that Eph-ephrin reactions may be adhesive when ephrin concentrations are low but repulsive when ephrin concentrations are high.

It is also possible that Eph receptors have a binding partner that is not an ephrin or that Ephs adhere to other cells without binding partners. Either finding would be extremely exciting, Kemp said, since neither has been suspected before.

Whether or not a novel mechanism waits to be discovered, "it's now apparent that Ephs and ephrins can produce repulsive or adhesive interactions," Kemp said. "It depends on the context."

Fred Hutchinson Cancer Research Center is a world leader in research to prevent, detect and treat cancer and other life-threatening diseases.