Sorting wheat from chaff
Image courtesy of Nadav Kupiec, Broad Institute
Tumors harbor an array of genetic misspellings, only a small fraction of which is likely to contribute to cancer. Bringing light to this influential minority requires ways of filtering vast amounts of information, as current DNA-scanning tools churn out massive lists, with no clear distinction between more and less meaningful mutations. In a groundbreaking new study, a team of scientists at the Broad Institute, Dana-Farber Cancer Institute, and Brigham and Women’s Hospital offers a paradigm for sorting the genomic wheat from the chaff.
Published in the June 14 issue of Cell, the researchers describe a multi-tiered approach aimed at unveiling the functional consequences of the genetic abnormalities found in cancer cells. By applying this approach, the scientists pinpointed a gene that is implicated in breast cancer. The gene, named IKBKE, was not previously suspected to play a role in the disease, highlighting the potential of genomic approaches to unearth novel findings. Importantly, IKBKE is among a special class of genes, called kinases, which are considered prime targets in the design of modern anti-cancer drugs.
“We are particularly excited about the early success of using an integrated genomic approach, because it represents a strategy for discovering cancer genes that can be applied to many cancer types,” said senior co-author William Hahn, a senior associate member at the Broad Institute and an associate professor of medicine at the Dana-Farber Cancer Institute and Harvard Medical School.
Like many areas of biomedicine, a major challenge in cancer research is to determine what goes wrong within cells to cause disease. Indeed, scanning cells’ DNA for abnormalities can yield important clues. Yet in cancer, the genome is riddled with errors, many of which have little or no effects. To exclude these red herrings, scientists require sophisticated methods for exploring how mutations fuel cancer growth. Such methods need to be deployable at a large-scale to match the rapid nature of ongoing cancer genomics efforts.
Toward this end, a scientific team co-led by Hahn undertook a three-pronged approach to identify human oncogenes. Oncogenes — a type of cancer-causing gene — are frequently mutated in tumors. And these mutations not only render the genes unusually active, they also serve as a kind of Achilles’ heel — a vulnerable spot that can often be exploited to kill cancer cells.
Putting the pieces in place
In the first phase of the project, Hahn and his colleagues developed an artificial system for coaxing normal cells to become cancerous, using a defined group of abnormal genes. The set-up takes its cue from a well-known oncogene called RAS, which dispatches signals from cell surfaces via a large, multi-gene network. Mutations in RAS, or in one of its many partner genes, are among the most common genetic abnormalities in human cancers.
The researchers turned their attention to two such partners — MEK1 and AKT. When these two genes were mutated in concert, they substituted for RAS, triggering a cascade of events that drove cells toward malignancy. But alone, neither was able to promote cancer, the scientists found.
In addition to shedding light on crucial parts of the RAS signaling network, this work also provided a new kind of testing ground. That is, by removing one of the two cancer culprits, the researchers could systematically introduce other genes to test their capacity to restore tumor formation. In this way, the scientists analyzed more than 350 kinases and kinase-related genes, and identified four potential oncogenes: IKBKE, DAK, TSSK6, and CSNK1E.
Back to DNA basics
Hahn and his colleagues then used genome-scanning technologies to determine if these four genes are ever altered at the DNA level in cancer cells. By searching the DNA of both laboratory cancer cell lines and tumor samples derived from cancer patients, the researchers zeroed in on one gene — IKBKE.
Surprisingly, the IKBKE gene was abnormal in as many as 40% of the breast cancer samples examined. Instead of the usual two copies, the gene was present at elevated levels, ranging from five to ten copies. Upon closer inspection, the researchers found that this phenomenon — a so-called “copy number” gain —involved more than just the IKBKE gene. A large region on chromosome 1 (designated 1q32), where IKBKE is housed, was also amplified.
Importantly, copy number gains of this sort are known to be one of the most common genetic alterations in breast cancer cells. This type of mutation is distinct from those that can occur within the known breast cancer genes, BRCA1 and BRCA2. Mutations in these genes are passed from parent to child, whereas IKBKE mutations arise in a woman’s own cells during her lifetime.
An Achilles’ heel
The researchers knew that increased IKBKE activity can lead to tumor formation and that the gene is frequently mutated in cancer cells — two defining features of an oncogene. But another key piece of the puzzle was still missing. Is the increased activity of IKBKE necessary to maintain cancer growth?
To address this question, the scientists turned to a collection of RNA-interference (RNAi) tools that can systematically reduce or “knock down” the activity of specific genes. The tools, which target nearly every gene in the human genome, were recently developed by The RNAi Consortium (TRC), a public-private collaboration including the Broad Institute as well as several other academic research institutions and life sciences organizations. Hahn and his colleagues used these reagents to test 1200 human genes, including IKBKE as well as nearly every other kinase.
Much to their surprise, the researchers found that reducing IKBKE activity, specifically in breast cancer cells with amplifications of chromosome 1q32, caused the cancer cells to die. But other types of cancer cells, which lacked the genetic defect, were unharmed. These results suggest that IKBKE acts as a sort of molecular crutch that breast cancer cells need to survive.
“IKBKE is particularly attractive as a potential target for cancer therapeutics, both because it is a kinase and because it is expressed in a large percentage of breast tumors,” said Hahn.
The study appearing in Cell was led by co-senior authors William Hahn of the Broad Institute, Dana-Farber Cancer Institute and Brigham and Women’s Hospital; Broad Institute director Eric Lander; Kornelia Polyak of the Dana-Farber Cancer Institute and Brigham and Women’s Hospital; and Thomas Roberts of the Dana-Farber Cancer Institute.
Other Broad-affiliated scientists who took part in the work include Lauren Ambrogio, Jesse Boehm, Ian Dunn, Ron Firestein, Levi Garraway, Jennifer Grenier, Heidi Greuhlich, Greg Hinkle, So Young Kim, Matthew Meyerson, David Root, Rhine Shen, Sarah Sjostrom, and Carly Stewart.