Scientists introduce a new way of identifying key cancer genes

A new study shows that a family of molecules called microRNA work together in single, well-connected networks to control many important functions in healthy cells, but that in cancer cells the networks are rewired and fragmented.

The research, led by scientists at the Ohio State University Comprehensive Cancer Center-Arthur G. James Cancer Hospital and Richard J. Solove Research Institute in collaboration with investigators at 11 other centers, introduces a new way of discovering cancer genes and identifies new miRNAs that can be used as targets for drug development and pinpoints possible new cancer-related proteins, says study leader Dr. Carlo Croce, professor of molecular virology, immunology and medical genetics, and director of Ohio State's Human Cancer Genetics program.

MicroRNA (miRNA) are tiny molecules discovered 10 years ago that control important cell functions, including growth, proliferation and differentiation. Abnormal miRNA activity plays an important role in cancer development.

The new study, published in the May 3 issue of the journal Genome Research, shows that the miRNA network in healthy cells, when mapped, resembles a family tree with dozens to hundreds of members. Each cell type has its own specific network, with particular miRNAs playing a more central role and serving as hubs within the network.

In cancer cells, however, the single network is replaced by subnetworks that usually include small detached clusters of two to six miRNAs.

"The presence of these small groups that exist outside the main network was completely unexpected," Croce says. "Some of these miRNA outliers are well known cancer genes, while the involvement of others in cancer was unknown."

Previous studies have measured differences in individual miRNA levels between cancer and normal tissues, says first author Stefano Volinia, assistant professor of biomedical informatics and of molecular virology, immunology and medical genetics.

"The molecules were always considered individually and, generally, unrelated to each other, because we did not know much about how miRNAs cooperate," Volinia notes. "In this study we show that key cancer genes can be identified on the basis of their relationships, rather than on their overactivity or loss."

To complete this study, the researchers studied miRNA expression profiles in 3,300 human cancer samples and 1,100 healthy tissue samples. These samples represented 50 different normal tissues and 51 cancer types, 31 of which were solid tumors and 20 leukemias and lymphomas.

"The size of our database enabled us to analyze coordinated miRNA activities and to build miRNA networks for different solid tumors and leukemias," Croce says.

In lung cancer, for example, the study mapped a subnetwork of 16 miRNAs plus eight groups of two or six miRNAs that are detached from the larger subnet. Lung cancer and other solid tumors showed more of these detached groups than did leukemias.

"Our findings suggest that we should reconsidered miRNA activity in cancer as the cooperative work of small groups of rebellious genes," Volina says.