Tumor growth requires abnormal cell neighbors

For some brain tumors, it’s not just what you know but who you know.

In trying to develop a mouse model of a genetic disorder that predisposes children to certain types of brain tumors, a University team discovered that tumors only developed when genetic abnormalities were present in all brain cells, not just in those that become cancerous.

The study was featured on the cover of the Dec. 15 issue of Cancer Research.

David Gutmann

“We are quite excited about this report as it represents the first model of this type of tumor,” said principal investigator David H. Gutmann, M.D., Ph.D., the Donald O. Schnuck Family Professor of Neurology.

“We’ve always assumed that cancer results from the loss of specific genes in a particular cell, but apparently that isn’t always the case. Our findings suggest that as in real estate, location is everything — a permissive environment may be the key to whether a tumor cell becomes cancerous or just sits dormant for a person’s entire life.”

According to the National NF Foundation, the genetic disorder neurofibromatosis 1 is the most common neurological disorder caused by a single gene.

The disorder can lead to a variety of complications including skin, spine and brain cancer. Up to 20 percent of patients with the disorder develop tumors in a type of support cell called an astrocyte along the optic nerve and optic chiasm, which transmit visual information from the eye to the brain.

Astrocytes that develop into tumors lack both copies of the Nf1 gene. So Gutmann’s team first developed genetically engineered mice in which all cells were normal except astrocytes, which lacked both copies of the gene. To the researchers’ surprise, the mice did not develop brain tumors.

Humans with the disorder are born with one normal and one mutated copy of the gene in all cells in their bodies. Gutmann’s team therefore hypothesized that genetic abnormalities in brain cells surrounding astrocytes might be essential for tumor formation.

To test this theory, the team developed mice with no functional copies of the gene in their astrocytes and only one functional copy in all other brain cells, a scenario identical to that of humans with the disease. Every mouse developed astrocyte tumors along the optic nerve or chiasm within the first 10 months.

According to Gutmann, understanding the events that lead to tumor growth is critical for learning how to predict — and hopefully prevent — tumors.

“It’s clear from our findings that tumors do not form simply by losing both copies of the Nf1 gene,” he explains. “If we figure out what external cues are necessary to trigger tumor growth, we could try to shut off that switch and stop tumors dead in their tracks without having to correct the underlying genetic defect.”

The potential for the mouse model used in this study to serve as a preclinical model of the disorder is enhanced by the team’s ability to detect tumors in very early stages using a powerful 4.7-Tesla MRI scanner and algorithms developed by Gutmann’s colleagues at the Mallinckrodt Institute of Radiology. Their techniques and equipment enable them to detect tumors the size of a piece of thread.

“We’re now beginning to detect these tumors even earlier using MRI,” Gutmann said. “I think we’ve gotten to the point where this mouse model can not only help us understand more about the cell biology underlying brain tumor development, but it also provides a tool for developing and evaluating better treatments.”