Duke Researchers Identify 10 Genes Believed To Protect Oxygen-Starved Cancer Cells

Led by Dr. Gregory J. Riggins, an assistant professor of pathology and genetics at Duke, the team sifted through 24,504 genes expressed in the oxygen-deprived (hypoxic) cells in Glioblastoma multiforme, which is a form of brain cancer. The researchers identified 10 genes, some of which can lead to the creation of new blood vessels that will connect to oxygen-starved tumors.

The findings appear in the Sept. 5 issue of the Journal of the National Cancer Institute. The research was done in collaboration with the Cancer Genome Anatomy Project in Washington, D.C., and funded by the National Cancer Institute and the James S. McDonnell Foundation.

"This is a crucial first step in understanding the complex interactions driving hypoxia response in tumors," Riggins said. "The potential for exploiting these genes is tremendous in terms of turning off the angiogenesis (the formation of new blood cells) of cancer and it may have significant use in other types of diseases as well, such as stroke and heart disease. This is a very early, very preliminary discovery.

The gene discovery is the very first step in a very long road that will lead to something that has clinical use." Previously, scientists thought angiogenesis was primarily influenced by a small set of genes including the vascular endothelial growth factor protein (VEGF), but Riggins suspects that the number of genes involved is significantly greater.

"All of these genes are turned on to a higher level than what has classically been studied as a potent inducer of angiogenesis -- VEGF. If any one of these genes turns out to be as potent or more potent than VEGF, then this is a fairly significant find. There is significant investment from the pharmaceutical industry to try and inhibit the function of this one particular gene," he said.

Other scientists had previously identified 10 genes that respond to hypoxia. The new Duke discovery adds another 10 to the mix of which three had never previously been identified, Riggins said. "These are completely novel genes that no one knows anything about," he said. "We are working now to determine what their functions are beyond being induced by hypoxia. What specifically are they trying to do? Blood vessel growth? Are they trying to maintain pH in the tumor or are they trying to protect the cell in some other way?"

One of the10 genes is similar to a gene protein called angiopoietin-1, Riggins said. It has a structure that suggests it could be an angiogenesis promoting gene. And another of the genes, carbonic anhydrase IX (CA9), is believed to be a useful marker to predict survival in some breast cancer patients.

In addition to Glioblastoma multiforme, the same genes were expressed in, and could have implications for, treating squamous cell carcinomas, breast, colon and lung cancers, the researchers said. "If we can disrupt the function of some of these protective genes, then maybe we can find ways to kill off the part of the part of the tumor that's traditionally has been very difficult to treat," Riggins said.

Tumors need a blood supply to grow, but when this blood supply and the oxygen it carries is cut, producing the oxygen-deficient known as hypoxia, the tumor shrinks. Most solid tumors develop hard-to-reach hypoxic regions as they grow, but some tumors, particularly Glioblastoma multiforme, continue to thrive under hypoxic conditions. The cells in the tumors can alter gene expression to produce new blood vessels (a process known as angiogenesis) with VEGF and other proteins. Radiation and chemotherapy are often ineffective in fighting hypoxic regions of tumors.

Because cancers are so complex, Riggins said targeting one gene isn't enough. Instead, the entire pattern of gene expression should be examined, and treatments formulated based on more than one gene. "Cancers can adapt rapidly to environmental challenges, which partially explains why single modality cancer therapy had not been successful as a multi-agent therapy. This work documents a more complete set of genes whose function may be to allow the tumor to adopt to hypoxia. More importantly, the products from such genes may be tumor-specific targets for therapy.

Based on the observation that many angiogenesis-related genes are hypoxia-induced, it is possible that successful targeting of the right combination of hypoxia over-expressed genes (HOGs) could result in the disruption of growth of blood vessels in tumors," the authors wrote in the article. "The next step is to gather more documentation as to precisely what the molecular function of these genes are in the tumor. We are currently synthesizing (replicating) the proteins of these genes and testing to see if alterations of these genes cause tumors to grow differently in mouse experimental models," Riggins said.

The study's co-authors are Anita Lal, a postdoctoral fellow in pathology at Duke; Mark W. Dewhirst, professor of radiation oncology at Duke; Zishan A. Haroon, research associate in radiation oncology at Duke; Albert J. van der Kogel, Johannes H. A. M. Kaanders and Hans Peters, Institute of Radiotherapy, University of Nijmegen, The Netherlands; Brad St. Croix, the Johns Hopkins Oncology Center, Johns Hopkins University School of Medicine; and Robert L. Strausberg, Cancer Genomics Office of the National Cancer Institute.

Riggins is the director of the Duke Brain Cancer Genomics Laboratory. The laboratory specializes in locating genes useful for the diagnosis and treatment of brain cancer, as well as understanding how these cancers develop.