New class of cancer-causing genes:
aberrant "microRNA" linked to human cancer

"These studies change the landscape of cancer genetics." Dr. Paul Meltzer.

PSA Rising, June 9, 2005. Scientists have discovered that a new class of genetic regulators called microRNAs can cause cancer. This may lead to new ways to classify, diagnose and treat some cancers.

microRNAs are sequences encoding tiny bits of genetic material contained in the DNA of plants and animals which control processes determining cell type and cell death.

A gene is a short piece of DNA. Each gene encodes a unique protein ("A large complex molecule made up of one or more chains of amino acids. Proteins perform a wide variety of activities in the cell"). Each protein performs a specialized function in the cell of animal bodies or plants.

Genes send and receive chemical messages. Molecular messengers 'translate' genetic signals into the chemical language of the recipient. Scientists have long thought of RNA as the intermediate 'messenger' between genes and proteins, because DNA is transcribed into RNA before being translated into protein.

But recently scientists have discovered that many genes are transcribed into RNA and never made into proteins. These 'non-coding' RNAs seem to act as regulators rather than messengers. They influence the expression of genes, for instance, by destroying coding RNA molecules so they cannot be translated into proteins.

This process was known to be important in normal human growth and development, but recent research shows that it can slow down or start up cancer.

For example, the Ras cancer gene is out of control in about 20 percent of cancers where it is over-expressed or activated by mutation. This spring, research in the laboratory of Assistant Professor Frank J. Slack at Yale University identified a new way that the Ras gene is regulated in lung cancer, presenting new possibilities for diagnosis and treatment.

"The let-7 microRNA regulates Ras by binding to the message for Ras and likely inhibits translation of the Ras protein," said Yale Cancer Center Assistant Professor Frank J. Slack. "The microRNA does not revert a mutated Ras to normal; instead it acts like a brake on an accelerated Ras."

In a project at University of Connecticut, scientists believe patients with androgen-independent recurrent prostate cancer may have mutations in a "microRNA target site in androgen receptor mRNA." They plan to test whether these mutations result in the up-regulation of androgen receptor expression and confer resistance to androgen ablation therapy.

Now researchers have found that microRNAs can trigger tumors. Research that appears in Nature this week shows that a type of non-coding RNAs called micro-RNAs can cause disease. And work by Dr. Gregory Hannon at Cold Sprin Harbor in New York shows that some microRNAs can accelerate the development of cancer when transplanted into mice.

Identifying these microRNA fragments may help detect even hard-to-diagnose cancer, researchers say.

New York's Cold Spring Harbor Laboratory found raised levels of microRNA activity were linked to aggressive forms of blood cancer.

A second study found different cancers can be identified by their highly distinct patterns of microRNA activity.

These discoveries could improve the ways cancers are classified, diagnosed and treated. "Everybody has been focused on understanding how proteins fit into cancer-causing pathways, but there is another regulatory pathway that controls not the protein product but its utilization," Greg Hannon of the Cold Spring Harbor Laboratory in New York told the Nature reporter.

Genes operate and are operated on by sending out and receiving chemical messages (a kind of chemical email). A gene is a short piece of DNA, which tells the body how to build a specific protein. Scientists have long thought of RNA as the intermediate 'messenger' between genes and proteins, because DNA is transcribed into RNA before being translated into protein.

But during the past few years, scientists have discovered that many genes are transcribed into RNA and never made into proteins. These 'non-coding' RNAs seem to act as regulators rather than messengers. They influence the expression of genes, for instance, by destroying coding RNA molecules so they cannot be translated into proteins.

This process was known to be important in normal human growth and development, but research that appears in Nature this week shows that a type of non-coding RNAs called micro-RNAs can cause disease as well. Hannon's work, for example, shows that some microRNAs can accelerate the development of cancer when transplanted into mice

Dr. Hannon at Cold Spring and a colleague at University of North Carolina, Chapel Hill, Dr. Scott Hammond, focused on a segment of human chromosome 13 that was known to be amplified, or over-expressed (i.e. present in excess) in several tumor types including B-cell lymphoma. The researchers observed that this DNA segment, referred to as the mir-17-92 cluster, has the potential to encode seven microRNAs.

To see whether excess expression of microRNAs encoded by the mir-17-92 cluster might be involved in cancer, the scientists first examined whether one or more of the microRNAs was expressed at abnormally high levels in four B-cell lymphoma cell lines in which the mir-17-92 cluster was amplified, compared to normal B-cells and to five leukemia and lymphoma cell lines in which the mir-17-92 cluster was not amplified.

The researchers found that indeed, five microRNAs encoded by the mir-17-92 cluster were overexpressed specifically in the B-cell lymphoma cell lines bearing an amplified mir-17-92 cluster.

Next, the scientists examined the expression levels of the mir-17-92 microRNAs in human tumor biopsies including 46 lymphomas and 47 colorectal carcinomas. They saw significant (greater than fivefold) overexpression of the mir-17-92 microRNAs in 65% of the lymphomas, with an average mir-17-92 microRNA overexpression of 10-fold in those lymphomas (and a high of 82-fold microRNA overexpression). In contrast, 15% of the colorectal carcinomas displayed greater than fivefold mir-17-92 microRNA overexpression.

" These findings suggested that mir-17-92 microRNA overexpression might contribute to human cancer, particularly to B-cell lymphoma but also to other forms of the disease," says Hannon. "And they gave us confidence that we might be looking at something that would be clinically relevant."

To test that idea directly, the researchers examined whether elevated expression of mir-17-92 or other microRNAs could speed up onset of cancer and/or decrease survival in a mouse model of B-cell lymphoma.

As expected, control animals developed B-cell lymphoma after about 2 months, and the overall survival of this group (n = 12) after 3 months was 75%.

In stark contrast, animals in which the mir-17-92 microRNAs were overexpressed experienced accelerated development of B-cell lymphoma (40 days compared to 2 months), and--significantly--none of the animals in this group (n = 14) survived after 3 months, compared to the 75% survival rate of the control animals in which the mir-17-92 microRNAs were not overexpressed.

Moreover, tumors induced in mir-17-92 microRNA-overexpressing animals consistently invaded organs outside the lymphoid compartment (including liver, lung, and kidney), and lacked the extensive "programmed cell death" or apoptosis observed in the control tumors and which helps keeps tumors in check. These findings indicate that overexpression of the mir-17-92 microRNAs can contribute to highly malignant tumors.

Collectively, the results of the study establish that microRNAs can function as oncogenes, leading the researchers to propose that such oncogenic microRNAs be designated "oncomiRs," with mir-17-92 being oncomiR-1.

"This is by no means a final answer about the role of microRNAs in cancer," says Hannon. "But it's the first definitive demonstration that microRNAs can act as oncogenes."

For more information, visit www.cshl.edu .