Science Editor

In one sense, cancer genes are a dime a dozen. There are literally thousands of genes that are associated with cancer, in the sense that their expression level is somehow different in normal and cancer cells.

But go looking for causal genes, and the ranks thin considerably. The list "gets dramatically shortened, to probably about 100 genes at the moment," Frank Slack told BioWorld Today.

And of those genes, the ones whose activity is truly necessary for tumors to keep on keepin' on is shorter still. "Ten to 15 genes have been shown to be required for the tumor to survive" – a list that includes oncology household names such as EGFR, ras and myc.

The tumor's dependence on constant activity of such genes has been termed oncogene addiction, which differs from the mere gene up-regulation in pretty much the way its name would imply: no matter what, the cancer cell's gotta have it. "If you target those genes . . . the tumors stop growing or regress and shrink.

Now, Slack and his colleagues Pedro Medina and Mona Nolde – all of them at Yale University – have added the first microRNA to that elite list; miR-21.

Their findings are published in the Aug. 8, 2010, online edition of Nature.

Slack and his team looked at miR-21 because it is overexpressed in most cancers where anyone has cared to look.

They engineered mice to contain an inducible transgene for miR-21: a transgene that was present, but whose expression can be blocked by antibiotics in the diet. Feeding the animals a regular diet turns expression of the gene on, leading to high levels of miR-21, while a dose of antibiotics with their chow turned the miR-21 gene off.

Using that model, Slack and his team showed that miR-21 is critical both for the initiation of cancer, and for the tumors to keep flourishing once they were established. "Only in mice where we turned on the gene did we see any tumors," Slack said. Such animals developed B-cell lymphomas, and were dead within a few months unless the scientists added antibiotics back to their diet.

When the expression of miR-21 was turned off, even mice with large tumors made comebacks. When the researchers stopped miR-21 in 10 animals with "conspicuous lymphoma," tumors regressed in every one of the animals, and all of them survived for at least three months – a time point at which all control animals who kept expressing miR-21 had died.

There are two ways in which tumor cells can disappear: They can either differentiate into regular cells that then no longer divide, or they can die outright. When Slack and his colleagues treated miR-21 deprived cells with anti-apoptotic drugs, those cells kept dividing. Apparently, miR-21 acts by preventing cell death in cancer cells.

Slack said his team is working on using the model "as a platform for developing anti-miR-21 therapeutics."

They also plan to engineer animals who express miR-21 only in specific organs, which they hope will enable them to study a wider range of cancers. With the current model, though the overexpression of miR-21 occurred throughout the bodies of transgenic mice – Slack and his team checked for and found high levels of miR21 in several organs, including brain tissue – the mice very specifically developed B-cell lymphomas.

The reason, Slack said, is "trivial: The animals die of lymphoma before they get a chance to get other tumors." Organ tumors in mice usually take three months to a year to form, and the transgenic animals died within three months of developing lymphomas.

"In a sense, we have an embarrassment of riches," Slack said. "The lymphoma is swamping out other things we would love to study."