Skip to main content
Premium Trial:

Request an Annual Quote

Cracking the Cancer Code


More and more, it seems, the effects of systems biology-related disciplines are being felt within the cancer research community. That could be a result of the tremendous amount of money that’s been poured into cancer research in the past few years, or perhaps the urgency with which researchers see the need to cure this complex group of diseases.

Whatever the case, that genomics, proteomics, and the rest of their 'omic cousins are now critical troops in the battle against cancer is an undeniable reality. In this third annual cancer special issue, Genome Technology speaks with scientists leading the fight — specifically, researchers whose advances have in the last 12 months made a considerable difference in the community’s understanding of cancer. From RNA interference to bioimaging to pathway simulation and beyond, the stories captured here provide a snapshot of the wealth of great science that’s going on in this area.

Tumor Suppressors
CSHL group uses shRNA to find key suppressor switch

For 30 years it’s been known to lurk in the hidden depths of the human genome. Now, thanks to emerging systems biology technology, its face has been unveiled.

Researchers at Cold Spring Harbor Laboratory have added a major piece to the puzzle of how tumor suppression works, after identifying a novel tumor suppressor that serves as a “master switch” for a network of cancer-preventing proteins. Their findings were published in the February issue of Cell.

Using a decade-old technology to investigate abnormalities typically found in cancerous cells, Alea Mills, a cancer geneticist and associate professor at CSHL, has identified CHD5 as an important tumor suppressor gene located in a region on chromosome 1, known as 1p36. Not only does CHD5, also known as chromodomain helicase DNA binding protein 5, enhance tumor suppression on its own, it also activates two important tumor suppressor cascades, the p53/p19 and the p16/Rb pathways.

“I was super-convinced we hit a critical region,” Mills says of engineering efforts that allowed her to both add and delete genes on the chromosome, which prevented and caused cancer, respectively.

Over the past 30 years, deletions in 1p36 have been found in many types of cancers, including brain, blood, and epithelial cancers, such as breast, colon, and prostate. This is the first study to link a specific gene on that locus to cancer, and the first to identify a chromatin-remodeling protein to function in tumorigenesis.

The first part of Mills’ study involved engineering mice that had either additions or deletions of segments of the 1p36 region. Using a technique called chromosome engineering — first pioneered in Allan Bradley’s lab and further advanced by Mills — she was able to cut, copy, and paste specific regions of the mouse genome. Utilizing this technique enabled Mills to analyze cancer from the start, rather than at end-stage tumor. Her team found that when they added extra copies of the 1p36 gene, tumor suppression was enhanced; when they deleted copies, cells became cancerous.

Isolating the gene from 52 potential ones in the 1p36 region — a total of 4.3 Mb of genetic material — was the next step. After selecting several candidates, they used RNAi technology to knock down each separate gene, in effect inhibiting the extra tumor suppression that the additional copy of 1p36 had afforded. The CHD5 gene was a standout in that it was the only one whose knockdown enabled cells to grow at a normal rate.

Mills then went on to explore how CHD5 might actually function as a tumor suppressor, and discovered that this gene “is sort of a master regulator of a tumor suppressor network.”

After a series of trials, Mills’ group was able to link deletion of CHD5 to both the p53/p19 and p16 tumor suppressor pathways. Using shRNA to knock down p53, they found that p53 expression was compromised when they deleted a copy of the gene, and enhanced when they added a copy. In other tests, they found that expression levels of p16 and p19, two tumor suppressors upstream of p53, were compromised in CHD5-deficient cells. When CHD5 is deleted, it can’t turn on p19, which can’t help to induce p53, which in turn stops cell division, repairs DNA, or initiates apoptosis — all of which prevent cancer.

Finally, to extend their findings to human cancers, they teamed up with Stanford University cancer researchers Hannes Vogel and Markus Bredel. Vogel and Bredel found that at least one of two copies of CHD5 was lost in 20 percent of human glioma tumors, which is a common, and usually deadly, form of brain cancer.

Cory Abate-Shen, co-director of the Prostate Cancer Program of the Cancer Institute of New Jersey, thinks the research is groundbreaking. “It is an exceptional piece of science — a truly gutsy approach to isolate tumor suppressors that has uncovered an important novel tumor suppressor and that provides new information on how this tumor suppressor works,” Shen says. She adds that the work provides a model for new ways to find tumor suppressor genes, which could be used to discover more of them.

While CHD5 has been pinpointed, its mechanism of action remains to be elucidated. Future work involves determining the structure of this protein, and how its specific chromatin remodeling capabilities function in cancer, Mills believes.

“We’ve really just started. We’re at the tip of the iceberg as to how [CHD5] functions as a tumor suppressor,” Mills says.


Genome Instability
Salk’s Karlseder elucidates link between cancer and aging

Aging and cancer just seem to go together. That’s why Jan Karlseder’s lab at The Salk Institute for Biological Studies researches these entwined processes to illustrate how an organism’s genome can become unstable.

Genomic instability arises from either losing telomeres from the end of a chromosome or from breaks in the DNA along the chromosome. After a cell has divided multiple times, its telomeres become critically short. Usually, the cell then dies or stops growing — a hallmark of aging. But sometimes it doesn’t, and that leaves the chromosome with broken ends. Other breaks sometimes arise in the middle of a chromosome and are supposed to be fixed by the DNA repair mechanism, but, again, these gaps are not always filled in. These broken ends of DNA, from telomere loss or internal breaks, may undergo translocation, and the shuffling of chromosome parts can lead to loss of gene function and cancer. “Pretty much every cancer cell has an unstable genome,” says Karlseder, who is an assistant professor in the regulatory biology laboratory.

Karlseder and his lab began by studying telomeres using Werner Syndrome as a model. People with this syndrome undergo premature aging beginning in their teens and also have elevated cancer rates. “Telomeres and Werner Syndrome have always been suspected to be involved with each other, but the general telomere shortening rate didn’t seem to be altered. That didn’t seem to be the reason for the aging syndrome. So we looked at individual telomeres and found sometimes they were missing completely,” Karlseder says. Lacking a telomere or two, they saw, leads to the accelerated aging in these patients. Now his team is looking into the role of redundancy to see if proteins involved in protecting telomeres have other jobs. “We want to know if these proteins play a role in intra-chromosomal replication as well … not only at the ends,” he says.

Recently, they also began integrating the model organism C. elegans into their work. In this nematode, they are looking for proteins involved in controlling replication, cell division, or the DNA damage repair machinery. “We are looking for factors that influence genome stability,” Karlseder says.

By applying information gleaned from genomic screens of C. elegans to cultured cells from people with Werner Syndrome, Karlseder and his lab hope to unearth how the early stages of cancer progress — and how a normal genome becomes unstable.


Pathway Modeling
GNS, UCSD collaboration links cancer to metabolic pathways

It was one of those collaborations that justify the cost of attending a conference. Iya Khalil, a cofounder of Gene Network Sciences, remembers that Steve Dowdy, a scientist at the University of California, San Diego, heard about her company’s pathway modeling technology at a conference. It seemed to fill a gap he had encountered in his own research, she says, and he approached GNS about starting a collaboration around his topic of interest — “trying to uncover the mechanisms that drove the G1 cell cycle,” Khalil says.

That partnership, which began about two years ago, led to a paper published recently in Molecular Systems Biology in which Dowdy and Khalil demonstrate a connection between the mammalian G1 cell cycle, a known contributor to cell reproduction and tumor growth, and a metabolic pathway.

Khalil, who was the PI of the project, says the GNS technology was used to reverse-engineer the biological mechanism from Dowdy’s real-life data. Dowdy supplied all of the starting data from his own experiments into the G1 cell cycle, and the GNS team used that to build a model of what was going on. GNS went back to Dowdy for additional information “to help constrain the model,” Khalil says. The data came from quantitative time-course experiments of several factors that were important in the cell cycle, she adds. “He had very good experimental capabilities.”

Once the data were rounded up and the GNS team satisfied with the model, Khalil says that actual computer processing was quick — “essentially a day on a supercomputer,” she says. The modeling process churns through the data looking for parameters that provide the best explanation for the data fed into it. “You essentially leave it to the global optimization algorithm,” Khalil says, “to come up with the best match or pathway that explains that data set.”

What the algorithm turned up was the need for a modifier, or a critical metabolic factor, that sparks the transition within the cell cycle from early to late phase — a key component of the onset of cancerous cell growth.

Dowdy and his team are now looking into promising biological factors that could be the metabolic kick-start needed to run the G1 cell cycle, Khalil says.


DNA Damage
Xu’s team hunts peptides to block dangerous kinase

As a seasoned radiation oncologist, Bo Xu knows well the clinic’s revolving door of cancer patients with tumors resistant to chemotherapy and radiation. “Because of my unique radiation oncology background, I know what the clinical needs are,” says Xu, senior biochemistry researcher at the Southern Research Institute. “We want to move from the bench to the bedside, so we’re trying to develop those translational projects.”

Xu’s lab is focused on the mechanisms that control cellular response to DNA damage agents. “The reason we’re interested in DNA damage is because DNA damage itself can cause cancer — and also DNA damage can be used to cure cancer,” he says. “If we know the mechanism of how cells respond to DNA damage, we may be able to manipulate the tumor response to chemotherapy and radiotherapy, and know how tumors start growing and progress.”

Research has shown that a number of cancers in humans are related to mutations that affect proteins involved in cellular DNA damage response. For example, mutations in DNA damage response genes ATM and the Fanconi Anemia gene have been linked to leukemia and lymphoma. Other damage response genes, such as p53, BRCA1, and BRCA2, can lead to ovarian and breast cancers.

Xu is currently studying ATM kinase, an enzyme produced by the ATM gene, responsible for the debilitating immunodeficiency disorder ataxia-telangiectasia. Patients with this mutation can also develop leukemia or lymphoma, but are often hypersensitive to radiation and cannot benefit from traditional cancer treatments, Xu says. Using this disease model, he studies how ATM kinase regulates cellular response to DNA damage.

So far, his team has identified several possible targets. “ATM is basically an enzyme with which you can modify downstream targets, and there are many famous proteins that are ATM targets, for example p53 and BRCA1,” says Xu. When ATM is activated, it can promote cell survival, but when blocked, tumor cells become more susceptible to radiation or chemotherapy. Xu and his team have designed several peptides that can block ATM activation. “Whenever you add those peptides to the cells, the protein enzyme cannot function, so then the cell becomes very sensitive to radiotherapy,” he says. “That’s the way we are developing therapeutic drugs for targeting DNA damage pathways.”

Xu says that the next step is to bring these biological concepts into translational studies. This will involve the use of high-throughput screening as a means to identify new targets for drug development. “When the drugs are available, we can use it in the in vitro setting or animal models, then a clinical setting,” he says. “So in five to 10 years, we should have some promising therapeutic drugs developed.”


At Scripps, cell-based assay pics help choose promising compounds

Evelyn Griffin’s work has changed over time, but the underlying interest has remained constant: understanding cancer. Now a research assistant in the discovery biology group at Scripps Florida, she is responsible for performing “the cell-based assays for triaging compounds to become potential [therapeutics],” she says. In essence, she’s the first step in a screening process to choose which compounds are promising enough to send to the next development stage.

As part of her work, Griffin took the image that would wind up winning GE Healthcare’s IN Cell Competition, a contest for best image generated by IN Cell technology. “The image is of fibroblasts which overexpress the gene src,” Griffin says. The gene, known to be involved in several kinds of cancer, is part of a test in the compound-screening triage phase at Scripps. “We can tell toxicity and efficacy of these drugs based on the morphology of these cells,” Griffin says.

In this particular image, Griffin has captured enough data to analyze cytoskeletal features, apoptosis, and nuclei characteristics.


RNA Interference
New pathway gives hope for higher chemo success rates

No one said that cancer cells aren’t crafty. When left to their own devices, they always seem to find a way to divide — and conquer.

However, researchers at MIT have discovered a new way to catch cancer cells off guard by targeting an intracellular signaling pathway that was previously only associated with the inflammatory response. A paper published in the February 13 issue of Cancer Cell details their findings.

Lead researcher Michael Yaffe, an associate professor of biology and biological engineering at MIT, showed that tumor cells, in fact, rely on the MK2 pathway for survival after chemotherapy. By knocking out this pathway, they showed that the cells became much more susceptible to chemotherapy treatment. On the other hand, normal cells were unaffected in the process.

If a drug was developed that could knock out the MK2 pathway, Yaffe says, “we could give that drug to patients and that would make their tumor cells more sensitive to chemotherapy and probably wouldn’t have much effect on normal cells.”

In fact, one such drug is already in clinical trials at the NIH. Called UCN-01, this anti-tumor agent interferes with the MK2 pathway.

“I suspect similar drugs will be coming through the pipeline in the next two to three years,” says Yaffe, who is affiliated with MIT’s Center for Cancer Research, the Broad Institute of MIT and Harvard, and Beth Israel Deaconess Medical Center.

When DNA is damaged, cells rely on specific intracellular signaling pathways to stop cell division and repair the DNA before it is copied into another cell. Two of these known pathways are the ATM-Chk2 and ATR-Chk1 pathways. In tumor cells — most tumor cells lack the p53 pathway, one that normally helps stop division when DNA is damaged — a third pathway is called into action. In the team’s experiments, researchers found that if they knocked out the p38MAPK/MK2 pathway in cells that had undergone chemotherapy, the cells could no longer defend themselves.

Yaffe used RNAi technology to knock out the MK2 pathway, and after dosing the tumor cells with the common chemotherapy drug cisplatin, found that the tumors were much more susceptible. When they moved from the lab to mice, treating MK2-deficient mice with very low doses of cisplatin, “those tumors melted away,” Yaffe says.

“If you’re a tumor cell, and you lose the p53 pathway, then you become dependent on the MK2 pathway,” he adds. “MK2 becomes critical to survive after chemotherapy.”

In light of the promising work at both NIH and various pharmaceutical companies already working on commercializing MK2 inhibitors, Yaffe says his lab’s next steps are aimed at searching for novel inhibitors of MK2. He has partnered with Matthias Gaestel, a professor of biochemistry at Hannover Medical School in Germany, to take the first step — creating mice that lack MK2 and see how they are more or less susceptible to cancer and chemotherapy.

Gaestel’s research focuses on the regulation of gene expression by post-transcriptional protein phosphorylation as well as the function of the ERK/MK5 signaling module. He believes that Yaffe has made a major finding in this work. In the future, says Gaestel, “we would like to learn whether MK3, a closely related sister kinase of MK2, is also involved in this regulation.”

Since MK2 is already known to regulate inflammation and the immune process, in addition to being called into action in tumor cells that have undergone damage due to chemotherapy, “small molecule MK2-inhibitors can be of double benefit fighting tumor growth and inflammation in parallel,” Gaestel says.


Stem Cells
Stowers scientists take aim at bad-seed stem cells

Stem cells are usually touted as potential cures for Alzheimer’s disease, Parkinson’s disease, or even spinal cord injuries. Some stem cells, however, may actually cause disease. More and more researchers think that tumors are seeded by cancer stem cells.

As a molecular and cellular biologist, Linheng Li, an associate investigator at the Stowers Institute for Medical Research in Kansas City, Mo., studies how these aberrant stem cells form. Motivated by other labs’ findings, Li and his lab shifted their focus from studying normal stem cells to these rare, one-in-a-million cancer stem cells. “This is an exciting field — it just started,” says Li, who is also an associate professor at the University of Kansas School of Medicine.

Cancer stem cells turn over more slowly than the other cells in a tumor — though still more quickly than normal stem cells. Since they divide slowly, conventional cancer treatment that homes in on the quickly growing tumor cells miss these stowaways. Then, when the tumor is eradicated, the cancer stem cells remain — and can seed new tumor growth.

But how these cells arise in the first place isn’t quite clear. To research the cascade of events that leads to the rise of a cancer stem cell, Li and his lab start with mouse knockout models, and then couple that approach with a variety of technologies. In particular, they focus on the Wnt and BMP signaling pathways that are involved in embryogenesis as well as cancer development since these pathways cross-talk — Wnt promotes stem cell production, while BMP inhibits it. “We try to block the pathway and ask the question, ‘What happened to the stem cells?’” Li says. To get a complete answer to this question, they use microarrays to compare gene expression profiles, immunostaining to look at any stem cell signal changes, and RNAi to knock down gene function.

And his lab has some promising results. In one of their mouse model systems, Li and his group saw that the loss of a tumor suppressor gene found in the Wnt pathway, called PTEN, increased the number of stem cells and changed the location in the intestine of these suddenly more mobile cells. These changes lead to de novo crypt formation, which in intestinal cancer is a precursor to polyp and tumor growth — indicating that the stem cells had become cancer stem cells.

Li is hopeful that his lab’s findings will have clinical applications. Cancer stem cells, he says, could be a key part of the future of cancer treatment. His lab is trying to develop a way to systematically identify cancer stem cells and their molecular signatures. This, he says, will aid in early diagnosis and provide a way to use small molecules to look for effective drugs.


Gene Expression
Compendia Bio culls transcriptome data for online analysis tool

For a company that officially launched last year, it’s no trivial matter that Compendia Bioscience’s lead product already has some 10,000 registered users. That’s because the product, Oncomine, an online database of gene expression information from public sources around the world, started out in academia and has been growing its fan base within universities and nonprofit research organizations for years.

With the growth in popularity came the growth into the commercial sector, and the opportunity to launch Compendia. A spinout of the University of Michigan and founded by scientists Arul Chinnaiyan and Dan Rhodes, Compendia officially launched early last year and is based on the Oncomine technology. “Early on we had this idea that it was much more powerful to look across the world’s gene expression data than just the data that our lab was generating itself,” says Rhodes, who serves as the company’s CSO. “Then, when there’s a new type of question to ask,” he says, it’s a simple matter to launch the query “having that data collected and standardized in the database.”

The idea of Oncomine, says Matt Anstett, Compendia’s vice president of scientific applications, is to “bring all this data into one central location” — that is, gene expression data from all sorts of studies all over the world. Compendia pulls together “all the gene expression data that’s out there in the public domain,” annotating it and contacting authors for supplemental information where possible, Anstett adds. “That allows us to set up additional analysis, [such as] a comparison that the author didn’t originally publish.”

That data and the resulting analyses are fed into a Web-based navigation tool that’s targeted at biologists and oncologists. Currently surveying 39 cancer types and subtypes, according to Anstett, Oncomine can enable researchers to, for instance, “set up a comparison between a patient subgroup and a larger group” and fish for particular cancer expression signatures of interest.

According to Rhodes, a new version of Oncomine will position the tool more specifically for scientists involved in translational research, helping with drug identification and perhaps stratifying patient populations.


The Scan

Self-Reported Hearing Loss in Older Adults Begins Very Early in Life, Study Says

A JAMA Otolaryngology — Head & Neck Surgery study says polygenic risk scores associated with hearing loss in older adults is also associated with hearing decline in younger groups.

Genome-Wide Analysis Sheds Light on Genetics of ADHD

A genome-wide association study meta-analysis of attention-deficit hyperactivity disorder appearing in Nature Genetics links 76 genes to risk of having the disorder.

MicroRNA Cotargeting Linked to Lupus

A mouse-based study appearing in BMC Biology implicates two microRNAs with overlapping target sites in lupus.

Enzyme Involved in Lipid Metabolism Linked to Mutational Signatures

In Nature Genetics, a Wellcome Sanger Institute-led team found that APOBEC1 may contribute to the development of the SBS2 and SBS13 mutational signatures in the small intestine.