By David N. Leff

When research oncologist Bert Vogelstein at Johns Hopkins University, in Baltimore, looked at colorectal cancer cells about a decade ago, he found that in 70 percent of them a novel protein was missing.

This high rate of loss made him and his colleagues suspect that what they were seeing was the expressed product of a tumor suppressor gene that had become mutated into its opposite, a tumor-promoting oncogene. They named this candidate gene DCC, standing for "deleted in colon carcinoma."

DCC thus joined a long and growing list, beginning with p53, of tumor suppressor genes gone bad. Very recent research indicates that it's gone quite a bit further than identification as just one more colorectal oncogene.

These new findings suggest that the DCC gene, and the DCC protein it encodes, have links to metastatic rather than primary tumor suppression, and to the development and degeneration of the human nervous system as well.

The chain of evidence in that linkage emerged from efforts to figure out just how the mutated DCC gene product killed that 70 percent of colon cancer cells. To be sure, the smoking gun was apoptosis — programmed cell death. But what pulled that apoptotic trigger?

The title of a paper in the current issue of Nature, dated Oct. 22, 1998, offers this explanation: "The DCC gene product induces apoptosis by a mechanism requiring receptor proteolysis." The article's co-senior authors are neuroscientist Dale Bredesen and enzymologist Guy Salvesen, of the Burnham Institute, in La Jolla, Calif.

The DCC gene encodes a transmembrane receptor called netrin-1. That ligand turns out to be an axonal-guidance molecule, which acts as a pilot in the embryonic spatial organization of the neurons in the brain and nervous system.

"I think that the identification of netrin-1 as a ligand to DCC," Salvesen told BioWorld Today, "shifted the focus of that gene from the observation that its loss as a tumor suppressor correlated with the majority of colorectal cancers to a possibility that its real biologic function was axonal guidance in the developing nervous system.

"When you lose netrin-1 expression," he said, "you may run into a state where you have neurodegeneration. So we're now following that aspect more than the cancer aspects." This fundamental neuronal function, he noted, "goes back in evolution not just to mammals, not just to vertebrates, but with homologs to the fruit fly, Drosophila melanogaster."

Meet The Third Known 'Dependence Receptor'

Then Salvesen introduced "the really exciting finding that DCC, as a receptor that kills cells, is acting as what we call a dependence receptor. So when we supplied recipient cancer cells with DCC's ligand, netrin-1 — either in soluble form or co-transfected with DCC — we saw the rescue of those cells from apoptotic death. It was the presence of the dependence receptor that overcame the cell-killing effect of DCC."

Salvesen said that it was his co-author Dale Bredesen's "investigation into the nature of the dependence receptor that will make him famous in the next few years."

"The initial dependence receptor," he said, "the one that is probably best known, is the receptor for the common neurotrophin growth factor (NGF). Cells that express that receptor must have NGF in order to survive. If you remove the NGF from neurons, or other cells that have been transfected with its receptor, they undergo apoptosis.

"So in a sense," Salvesen suggested, "DCC was analogous to the neurotrophin receptor, except of course using different ligands, but the principle is the same. Such receptors create cellular states of dependence on their respective ligands by inducing apoptosis when unoccupied by ligand, but inhibiting cell death in the presence of ligand."

That principle obtains as well in the third dependence receptor thus far identified.

"One sees the same kind of effect with the androgen hormone and the androgen receptor," Salvesen said. "When that receptor is expressed in cells in the absence of the androgen ligand, it will cause apoptosis and kill the cell. But in the presence of androgen, there is a rescue.

"In other words," he explained, "these dependence receptors form an analogous set of proteins that exist in cells, and they addict the cells to the presence of the receptor. That would be useful in the development of the nervous system. They are going to specify the ability of the neurons to respond to, or die In the presence of, the correct neuronal developmental signals."

Salvesen tempered this interpretation by saying, "It's still a phenomenological state we're at right now, and the mechanism's only beginning to be explored."

Tumor Suppressor — Yes Or No?

As for DCC's traditional role, Salvesen said, "We're not quite sure that it really is a tumor suppressor."

To test the tumor suppressor hypothesis, other investigators a few years ago raised knockout mice lacking the DCC gene.

"If that DCC was a tumor suppressor, then homozygous mice wouldn't have anything," Salvesen pointed out. "But if you knocked it out in heterozygous animals, which carried one good copy of DCC and one bad copy, they should therefore have an increase in the rate of colorectal cancer.

"The idea with the tumor suppression model was you'd need to knock out both copies of the suppressor. Here you supply mice with only one copy of the tumor suppressor, so it should have a higher spontaneous rate of colonic cancer — but it didn't.

"What we suspect on the basis of this is that loss of DCC expression in itself is not transforming, but rather than being a tumor suppressor, it may be a metastasis suppressor. Because we would predict now that since DCC is a dependence receptor, the cells will be able to respond to its loss only in the context of metastasis.

"That is, now they become independent of their ligand for sustenance. Previously they were dependent on netrin's existence. Now they can travel without their ligand. And in that context one wouldn't necessarily expect to see an increase in the number of cancers in heterozygous mice.

"Obviously," Salvesen concluded, "it's easy for us to say this, but experiments in terms of metastatic expression have to be done." *