FINDINGS:Researchers have found that a particular gene is central to the brain cancer glioblastoma and will either fight the tumor or, conversely, help the tumor advance, depending on the tumor's genetic makeup.

RELEVANCE:These findings are relevant for the emerging field of personalized medicine. Glioblastoma is a highly aggressive type of cancer for which treatments remain extremely limited. Researchers have long assumed that the gene in question, STAT3, only acts as a tumor inducer, and so have been developing therapeutics that inhibit STAT3. But if STAT3 actually fights tumors in a subset of these cancers, such therapies would do more harm than good. These results may change the way researchers approach not only glioblastoma but other types of cancers as well.

PRINCIPAL INVESTIGATOR:Azad Bonni, Associate Professor of Pathology, Harvard Medical Schoolhttp://www.hmcnet.harvard.edu/pathol/labs/bonni/index.html

MULTIMEDIA:Video interviews with Azad Bonnihttp://hms.harvard.edu/public/video/ab1ybc80.mov http://hms.harvard.edu/public/video/ab2ybc80.mov

*See end of release for image links

CITATION: Genes and Development, Volume 22, Issue 4: February 15, 2008

Newswise — Perhaps the only positive spin one can put on the brain cancer glioblastoma is that it's relatively uncommon. Other than that, the news is bad. It is nearly always fatal, it tends to strike people in the prime of their lives, and the limited treatment options have changed little over decades. It's no wonder then that many researchers are determined to find new ways treat this poorly understood type of cancer.

One approach focuses on a gene called STAT3. In several tumors, STAT3 takes the role of an oncogene, that is, a gene whose normal functions are derailed and, as a result, becomes a driving force in a tumor's development. Clearly then, blocking STAT3 would deal a major blow to such tumors. But a new study led by a team at Harvard Medical School has found that STAT3 isn't always the villain. While it does behave as an oncogene in certain types of glioblastoma, in others it becomes what's called a "tumor suppressor gene," a type of gene often responsible for keeping the renegade cancer cells in check.

In other words, the same gene in the same cancer can play a completely different role from one person to the next, depending on genetic nuances between individuals. The results appear online February 6 in Genes and Development.

"This discovery lays the foundation for a more tailored therapeutic intervention," says Azad Bonni, an associate professor of pathology at Harvard Medical School, and senior author on this study. "And that's really important. You can't just go blindly treating people by inhibiting STAT3."

When most people think of brain cells, they think of neurons, those cells whose electric signaling gives rise to our consciousness. But another class of brain cells called astrocytes (named after their uncanny resemblance to stars) actually outnumber neurons ten to one. Despite their name, astrocytes play a less glitzy role than neurons do. Typically, they're support cells, involved with functions such as providing nutrients to nerve tissue and repairing scars. However, nearly all brain cancers occur in astrocytes, or in the neural stem cells that generate astrocytes.

Bonni, a neurologist and neuroscientist by training, decided to investigate the genetic etiology of glioblastoma by studying whether certain regulatory genes that control the generation of astrocytes during normal development also play a role in these tumors. The logic here is simple: since disease is often the breakdown of a normal biological process, the more we understand how cells get it right, the more we understand what can go wrong. And since STAT3 is a key gene that turns neural stem cells into astrocytes during normal development, what is its role in glioblastoma?

Bonni and two lead authors, Núria de la Iglesia and Genevieve Konopka, in collaboration with investigators in the laboratory of Ronald DePinho at the Dana-Farber Cancer Institute, began by genetically manipulating mouse astrocytes, then placing them into a second group of mice whose immune systems had been compromised. The findings surprised them. Taking advantage of previously published data, the researchers looked closely at how two genes, EGFR and PTEN—whose mutated forms are associated with glioblastoma—affect the function of STAT3 in astrocytes. Bonni's group found when EGFR is mutated, STAT3 is an oncogene; with a PTEN mutation, STAT3 is a tumor suppressor.

"EGFR, in its normal state, is a transmembrane receptor, usually performing its functions at the cell surface," says Bonni. "However, when it's mutated, we find it in the cell's nucleus interacting with STAT3—and turning it into an oncogene. STAT3 itself is not mutated or damaged. It's the process of regulating STAT3 that gets damaged."

With PTEN, it's a completely different story. PTEN is itself a tumor suppressor gene. When PTEN becomes disabled in astrocytes, these potential tumors still have STAT3 standing in their way. This is because STAT3 acts as a tumor suppressor normally in astrocytes. However, as more PTEN becomes disabled, a cascade of molecular events is set in motion with the express purpose of inhibiting STAT3 function and thus turning the tide in the cells toward tumor formation.

The researchers confirmed these findings in human glioblastoma tumors as well.

"The belief that STAT3 can only be an oncogene has been a pretty entrenched dogma in the field," says Bonni, "so we performed many, many experiments to make sure this was correct. It took some very persistent investigators in my lab to get the job done. As a result, we're convinced of our data."

While glioblastoma tends to be uncommon, STAT3 has also been implicated in prostate and breast cancers, so these results may translate to other types of tumors as well.

In addition, the findings contribute to the growing body of evidence for "personalized medicine," showing that many types of cancers contain subgroups that require different treatments.

This research was funded by the Stewart Trust of Washington, D.C., the Armenise-Harvard Foundation, and the Carolyn and Peter Lynch Research Fund.

Written by David Cameron

*IMAGES: 1-http://ecommons.med.harvard.edu/ec_res/nt/8CADCFEC-6ED6-4377-B3ED-6DEEC5FF14EE/ab1ybc80.jpgStat3 as oncogene: When EGFR—a cell-surface receptor—mutates in glioblastoma, the Stat3 protein becomes oncogenic and promotes tumor growth. The image on the left shows EGFR-mutated cancerous tissue in which Stat3 is active and thus advancing tumor activity. The right-hand image shows the same tissue without Stat3 activity, resulting in a healthier sample. (image courtesy of Harvard Medical School)

2-http://ecommons.med.harvard.edu/ec_res/nt/9B47863D-8A0E-4853-8B2E-16D6293AFA92/ab2ybc80.jpg Stat3 as tumor suppressor: In other subtypes of glioblastomas, Stat3 functions as a tumor suppressor gene. In the tissue samples above, the cancer has lost the tumor-suppressor PTEN, leaving that function almost entirely up to Stat3. In the left-hand image, an active Stat3 holds the cancer at bay. But when Stat3 is lost, as shown in the image on the right, the tumor grows. (image courtesy of Harvard Medical School)

Full citation:

Genes and Development, Volume 22, Issue 4: February 15, 2008

"Identification of a PTEN-regulated STAT3 brain tumor suppressor pathway"

Núria de la Iglesia(1), Genevieve Konopka(1,2), Sidharth V. Puram(1,3), Jennifer A. Chan(4), Robert M. Bachoo(5), Mingjian J. You(5), David E. Levy(6), Ronald A. DePinho(5), and Azad Bonni(1,2,3)

1-Department of Pathology, Harvard Medical School, Boston, MA2-Program in Neuroscience, Harvard Medical School, Boston, MA3-Program in Biological and Biomedical Sciences, Harvard MedicalSchool, Boston, MA4-Division of Neuropathology, Department of Pathology, Brigham and Women'sHospital, Boston, MA 5-Department of Medical Oncology, Center for Applied Cancer Science of theBelfer Institute for Innovative Cancer Science, Dana-Farber Cancer Institute, and Department of Medicine and Department Genetics, Harvard Medical School, Boston, MA6-Department of Pathology and Department of Microbiology, New York University School of Medicine, New York, NY

Harvard Medical School (www.hms.harvard.edu) has more than 7,500 full-time faculty working in 11 academic departments located at the School's Boston campus or in one of 47 hospital-based clinical departments at 17 Harvard-affiliated teaching hospitals and research institutes. Those affiliates include Beth Israel Deaconess Medical Center, Brigham and Women's Hospital, Cambridge Health Alliance, Children's Hospital Boston, Dana-Farber Cancer Institute, Forsyth Institute, Harvard Pilgrim Health Care, Joslin Diabetes Center, Judge Baker Children's Center, Immune Disease Institute, Massachusetts Eye and Ear Infirmary, Massachusetts General Hospital, McLean Hospital, Mount Auburn Hospital, Schepens Eye Research Institute, Spaulding Rehabilitation Hospital, and VA Boston Healthcare System.

MEDIA CONTACT
Register for reporter access to contact details
CITATIONS

Genes and Development (15-Feb-2008)