Signaling System Found that Halts the Growth of a Childhood Brain Cancer

A discovery by St. Jude Children’s Research Hospital scientists suggests a safer way to treat medulloblastoma, a rare but often fatal childhood brain tumor. The group found that one of the brain’s signaling pathways inhibits the growth of the highly aggressive cancer cells.

The researchers discovered that three proteins, designated BMP2, BMP4 and BMP7, halted the growth of medulloblastoma tumors and induced the malignant cells to develop into normal neurons.

“We think we have identified a pathway that can be used to prevent tumor formation and a potential target for therapy,” said Martine F. Roussel, Ph.D., a member of the St. Jude Department of Genetics and Tumor Cell Biology. A report on this work appears in the March 15 issue of Genes & Development. Roussel is the paper’s senior author.

Medulloblastoma occurs in the cerebellum, which is located in the lower, rear part of the brain. This cancer strikes about 350 young children in the United States annually. Although treated patients have an overall five-year survival rate of 70 percent, conventional therapies combining surgery, irradiation and chemotherapy frequently lead to permanent neurocognitive impairment.

Several research teams are seeking to decipher the intricate signaling mechanisms that govern the proliferation of cells called granule neuron progenitors (GNPs). These cells go on to develop into neurons in the cerebellum during the first year of life. But the disruption of this differentiation process can trigger medulloblastoma.

“We were interested in whether there were signals that inhibited tumor formation,” Roussel said. “And if there were, which ones were they? Could they be used to identify new therapeutic targets?”

Previous research had shown that spurring GNPs to differentiate into neurons requires that BMPs bind to a set of receptors on the cell surface. This binding results in blocking the activity of a signaling pathway triggered by another molecule called Sonic hedgehog.

“What was not known, and what we now find, is that the effect of BMPs on normal GNP cells is almost exactly mimicked in GNP-like tumor cells,” Roussel said.

In cell culture experiments, her group found that BMPs rapidly cause the degradation of a protein called Math1, which occurs in dividing GNPs, but not in non-proliferating neurons. Twelve hours after BMP treatment, researchers could detect no Math1 and cell growth soon stopped.

The exact way Math1 works remains unknown. However, in mice the protein is vital to the formation of a normal brain. Mice genetically altered so they did not carry the gene for Math1 failed to develop cerebellums.

The St. Jude team also performed gene transfer experiments in mice to test BMPs as a possible medulloblastoma treatment. Using a genetically altered virus, scientists inserted the BMP gene into the cancer cells and showed that the transfer not only halted tumor growth, but induced the cancer cells to change into neurons.