Cancer’s Ignition Switch
Conquest - Spring 2008
By Scott Merville
Long before cancer threatens health, long before it’s even detectable as cancer, normal cells stealthily are going bad. Identifying these precancerous changes during “the invisible stage” of a cell’s path to malignancy could make a huge difference in cancer prevention and treatment.
Bogdan Czerniak, M.D., Ph.D., and his colleagues have developed a unique map that pinpoints an organ’s normal, transitional and cancerous cells in a way that also depicts how cells arrive at a cancerous destination.
M. D. Anderson Pathology Professor Bogdan Czerniak, M.D., Ph.D., and colleagues are exposing this journey with a unique map — one so complex and finely detailed that it pinpoints an organ’s normal, transitional and cancerous cells in a way that also depicts how cells arrive at a cancerous destination.
The key finding from a decade of bladder cancer research is the existence of what Czerniak calls “forerunner genes,” whose inactivation paves the way for the silencing of tumor-suppressing genes, thus jump-starting cancer development.
“Forerunner genes are the ignition key that starts the engine of carcinogenesis,” Czerniak says. While the transition of normal cells into cancer has long been believed to involve activation of oncogenes and silencing of tumor-suppressing genes, forerunner genes drive the initial growth of precancerous cells.
“Discovery of these genes opens an entirely new field of investigation to identify biomarkers for the early identification and prevention of cancer,” he says. They also provide a new set of potential targets for therapies to prevent or treat cancer.
His technique, called whole organ histologic and genetic mapping, superimposes maps of genetic variation over a detailed map of the organ’s terrain as revealed by microscopic study of its tissue, or its histology.
Early detection, bladder cancer and beyond
Pathology Research Scientist Tadeusz Majewski, Ph.D., together with Postdoctoral Fellow Joon Jong, M.D., performed the high-resolution mapping of the 13q14 region containing the model tumor suppressor gene RB1, which provided critical information for the concept of forerunner genes.
Czerniak’s mapping technique and findings are the first of their kind, says T. Sudhir Srivastava, Ph.D., chief of the Cancer Biomarkers Research Group, Division of Cancer Prevention of the National Cancer Institute (NCI).
“Identifying genes involved in precancerous development has been an arduous task, primarily for lack of a systematic approach to discovering them and the non-availability of quality tumor specimens for discovery,” Srivastava says. “Dr. Czerniak has overcome these difficulties by utilizing the resources available at M. D. Anderson and employing the gene-mapping expertise of his group to uniquely characterize chromosomal regions involved in genomic imbalances, particularly those involved in progression of precancerous conditions to clinically aggressive bladder cancer.
“These findings will accelerate development of clinically useful biomarkers for the early detection, surveillance and clinical management of bladder cancer,” adds Srivastava, who leads NCI’s Early Detection Research Network, which partially funds Czerniak’s work.
There’s good reason to believe the team’s research will apply to other cancers. Lung and oral cancers are similar to bladder cancer, Czerniak notes. All are cancers of the epithelium, the tissue that lines the surfaces and cavities of the body’s organs. Epithelial cancers, or carcinomas, make up 80% of all cancers.
They have found that two forerunner genes, that lead to the development of bladder cancer when they are silenced, also are quieted to varying degrees in lung, breast, blood and common pediatric malignancies.
It’s hard to be first
The intense methods used in bladder cancer provide a model approach to address other cancers. There are new forerunner genes, new keys to the ignition, to be found.
At first, Czerniak expected to discover more crucial parts to the engine — new tumor-suppressors that are damaged or shut down, allowing cancerous cells to flourish. Instead, experimental data kept guiding Czerniak’s team back to the genomic neighborhoods of known tumor-suppressing genes, but not to the suppressors themselves.
Following the evidence in bladder cancer specimens, the team narrowed its search to several genes located near — one is actually inside — an important common tumor-suppressing gene known as RB1, which is shut down in many cancers. Researchers then demonstrated that two of these genes are silenced before RB1 is stifled.
These findings and the techniques used to reach them were so cutting-edge that publication took an agonizing four years. “Our initial publication was rejected by everyone, all of the major journals,” Czerniak says. “We hope that this is over.”
Reviewers at the prestigious Proceedings of the National Academy of Sciences (PNAS) finally embraced the forerunner gene paper enthusiastically and it was published last year. “But that paper is just the tip of the mountain,” he says.
Three Steps to Ignition
Inactivation of forerunner genes leading to the silencing of a major tumor-suppressing gene, RB1, leads to the establishment and proliferation of bladder cancer.
Here’s how this happens:
First: Everyone has two versions of each gene, one from each parent. Initially, one version of both the forerunner and the RB1 gene is inactivated by large deletions of DNA. The deletions in RB1 and its neighboring region are seen in about one-half of all epithelial bladder cancers.
Second: The remaining forerunner gene is inactivated by hypermethylation, the attachment of methyl chemical groups to spots on the gene that shut it down. This process is associated with expansion of low-grade tumors.
Third: The remaining RB1 gene is inactivated, most commonly by a mutation, a step associated with advancement to high-grade tumors.
‘The most precious piece of scientific work’
Jim Crawford, M.D., Ph.D., editor of Laboratory Investigation, a member of the Nature Publishing Group, will publish more of the mountain, devoting about one-half of the July issue to a “megapaper” that covers the team’s research to date.
“We’ve basically cleared out the issue for this paper,” says Crawford, who also is professor and chair of the University of Florida College of Medicine’s Department of Pathology, Immunology and Laboratory Medicine. Devoting that much space to a single paper is highly unusual, but Crawford is convinced it’s worthwhile. Czerniak notes that Crawford was one of the few editors who understood and supported the team’s early work. After the PNAS paper was published, the two discussed publication of a comprehensive article in Laboratory Investigation.
Crawford agreed to accommodate one huge paper, including supplemental data and figures used with permission from the PNAS paper. “A person should be able to read this paper and know the whole story,” Crawford says.
The resulting paper was too large to transmit electronically and had to be sent via FedEx. “When I received this paper I had the feeling I was holding the most precious piece of scientific work that I would ever touch in my career,” Crawford says.
Czerniak notes this research is a massive undertaking, involving multistep genetic screenings, followed by validation studies of initial findings, which required plenty of bladder, blood and urine specimens. Studies of gene expression, sequencing and a regulatory chemical process called methylation, along with functional studies of what candidate genes do and epidemiological analyses were required in the final steps of forerunner gene identification.
“Our collaborators, both inside and outside the institution, are superstars in their fields,” Czerniak says.
Faculty from six M. D. Anderson departments contributed to the Laboratory Investigation paper, and over the years, Czerniak’s team has collaborated with 15 departments. Internationally known researchers from Baylor College of Medicine in Houston, Creighton University in Omaha, Neb., The University of Texas Southwestern Medical Center in Dallas, the International Institute of Molecular and Cell Biology in Warsaw, Poland, and the Sunnybrook and Women’s College Health Science Center in Toronto made major contributions.
Did You Know?
The greatest risk factor for bladder cancer is smoking. Smokers are more than twice as likely to get bladder cancer as nonsmokers. Smoking causes nearly one-half of the deaths from bladder cancer among men and slightly more than a quarter in women.
Source: American Cancer Society
Mapping the terrain
It all starts with the surgically removed cancerous organ. In this case, it’s the bladder. Microscopic analysis of the entire organ creates a map of the bladder’s terrain. Areas of normal tissue are noted along with areas of precancerous plaques and lesions, and the tumor or tumors.
Taking the forerunner gene paper as an example, the team mapped the tissue of five bladders with cancer in the organ’s lining.
Next, they employed 787 DNA markers to identify chromosomal regions that display genetic variations of interest. By superimposing the low-resolution map of genetic variation over the geographic map of the organ’s tissue, they identified regions associated with two different types of precursor lesions.
In one type, a genetic change is obvious, but unaccompanied by notable change in a cell’s appearance or composition — its phenotype. In the second case, striking genetic change was accompanied by signs of cellular abnormality called dysplasia and development of carcinoma in situ, cancer that remains limited to its tissue of origin. Dysplasia and carcinoma in situ progress to higher grade, invasive cancer.
Additional analysis narrowed the chromosomal areas to portions of six chromosomes. These six sites were confirmed by testing multiple markers of genetic loss found in those chromosomal regions in the urine and blood of 32 bladder cancer patients and 31 disease-free patients with a history of bladder cancer. Genetic losses from at least one of the six regions were found in 98% of patients. In 82% of the cases, two to five chromosomal regions were involved.
“This told us that genetic losses in those regions were frequent in bladder cancers and that those regions may harbor genes critical to cancer development,” Czerniak says.
The team then chose the 13q14 region on chromosome 13, which they knew harbored the tumor-suppressor RB1 gene, for high-resolution genetic analysis to identify candidate genes. The initial mapping data indicated that inactivation of RB1 wasn’t associated with growth advantage of early in situ precancerous lesions and suggested that other genes in the region may be involved.
Sangkyou Lee, Ph.D., a research scientist in the Department of Pathology, and others on the team have a vital role in analyzing and understanding the function of forerunner genes.
Their earlier organ-wide genomic analysis employed microsatellite markers. These are very short stretches of repeat base pairs of nucleotides, which are the building blocks of DNA. For a more refined analysis of genes around RB1, the team used single-nucleotide polymorphisms, or SNPs, as markers. SNPs are points in the genome that vary by a single DNA nucleotide.
Researchers examined 92 SNPs mapping to a region around the RB1 gene in 84 paired samples of bladder tumors and blood DNA. This high-resolution genetic analysis pointed to the same section of the chromosome that whole organ histologic and genetic mapping had identified with expansion of abnormal cells.
Three genes in that segment were targeted for additional analysis. Two were found to be inactivated before RB1 loses its function during tumor development.
Researchers found that a neighboring gene called ITM2b was silenced by methylation, which is the attachment of methyl chemical groups to places on the gene that shut the gene down. In bladder cancer tumors and cancer cell lines, this gene was methylated 40% of the time. Methylation is an epigenetic effect, meaning it regulates gene expression without changing a gene’s DNA.
The smoking connection
A gene known as P2RY5 located inside a portion of the RB1 gene was affected by a number of single-nucleotide variations. A case-control study of one of the gene’s variant forms was conducted using blood DNA from 790 bladder cancer patients and 712 controls matched for age and gender. The study, conducted by Xifeng Wu, M.D., Ph.D., professor in M. D. Anderson’s Department of Epidemiology, found that the specific variation was present in 2.78% of patients.
Wu’s study also found that every single patient with this P2RY5 polymorphism who also smoked developed bladder cancer. Smoking is the most important risk factor in bladder cancer and thought to be involved in one-half of all cases.
Working with Henry Lynch, M.D., an expert in cancer genetics and professor and chair of Preventive Medicine at Creighton University School of Medicine, the team followed up earlier evidence that P2RY5 mutations might be inherited variations — what scientists call germline mutations — as opposed to variations caused by damage to existing DNA.
Genetic analysis of a family in Lynch’s research that has been plagued by a variety of cancers for generations showed that their tumors had only mutant P2RY5 genes. The genetic transmission pattern in the family indicated an association between inherited mutations of P2RY5 and cancer development.
“It was whole-organ mapping that led us to these inherited P2RY5 mutations,” Czerniak notes.
Exciting times ahead
This exhaustive, detailed scrutiny of a portion of one chromosome is what’s in store for the five other chromosomal regions identified by the whole organ map. In fact, the entire genome will get the high-resolution treatment.
As a urologist and member of Bogdan Czerniak’s research team, H. Barton Grossman, M.D., and others see the discovery of forerunner genes as providing new targets for treating, or even preventing, cancer.
The whole organ genetic and histologic mapping procedure is a useful model for hunting down forerunner genes in other types of cancer, particularly because many cancers share damage or inactivation of common tumor-suppressing genes.
The forerunner genes identified in bladder cancer were analyzed for their expression, methylation and sequence in 62 cell lines derived from major groups of common cancers. Of the cell lines tested, forerunner gene expression was reduced in 63% of the cases and ITM2b was methylated in 42%.
While forerunner gene down-regulation was identified in lung, breast, blood and pediatric malignancies, the genes were strongly expressed in colon and liver cancers. This makes sense, researchers note, because those two cancers don’t rely on inactivation of RB1 to thrive.
Czerniak remains excited by what’s to come. “It took us 10 years to get where we are now. With the new high-throughput technology now available, we’ll complete a high-resolution genetic map of the entire genome for bladder cancer in the next two or three years.”
Joe Gray, Ph.D., director of the Life Sciences Division of the Lawrence Berkeley National Laboratory, which is managed by the University of California, notes that researchers have made great progress identifying genomic and epigenetic aberrations that drive the initiation, invasion and metastasis of cancer.
“Dr. Czerniak and colleagues have now added an important three-dimensional histological context to cancer genomics that guides the identification of initiating events and promises to provide insights into how specific genomic aberrations contribute to distinct stages of tumor progression,” Gray says. “This approach should contribute substantially to our understanding of the functions of epigenetic and genomic aberrations in cancer.”
Conquest - Spring 2008
Download pdf version of Spring 2008