Master gene linked to childhood brain tumor

The master gene Math1 is involved in the genesis of the most common childhood brain tumor, medulloblastoma, said researchers from Baylor College of Medicine (www.bcm.edu) in a report in the current issue of the journal Science.

The finding provides a new treatment target in the deadly disease that most commonly affects children and young adults, said Dr. Huda Zoghbi (http://www.bcm.edu/pediatrics/index.cfm?Realm=99992437&This_Template=huda_zoghbi, http://www.bcm.edu/labs/zoghbi/index.html), the report's senior author and professor of molecular and human genetics, pediatrics, neurology and neuroscience at BCM.

Math1 (also known as Atoh1) is important for making a critical population of cells called granule neurons in the cerebellum, which controls important motor functions such as balance for walking and standing. Granule neurons are the largest population of neurons in the brain. They come from a small area at the top of the hindbrain. The precursors of these neurons start dividing there and then go to the cerebellum where they divide over and over to make hundreds of millions of cells.

Medulloblastoma, which affects an estimated 1,000 people in the United States every year, arises in the hindbrain and is thought to come from this dividing population of cerebellar cells.

Zoghbi and her team members, including Dr. Adriano Flora (http://www.bcm.edu/labs/zoghbi/Lab_members_info/flora.html), a postdoctoral associate in her laboratory, found that Math1 was critical to this activity, and they wondered what role it played in the tumor. Using a special technique, they removed Math1 in this brain tissue and found that the neurons stop dividing and differentiate into cells that cannot divide.

"That told us Math1 was important for continuous proliferation," said Zoghbi, who is also a Howard Hughes Medical Institute investigator and director of the Jan and Dan Duncan Neurological Research Institute (http://www.texaschildrens.com/Web/TCNRI/TCNRI.html) at Texas Children's Hospital (www.texaschildrens.org).

They then sought to determine Math1's role in medulloblastoma. When they activated a special pathway called sonic hedgehog signaling in mice, they found that the animals developed medulloblastoma. However, if they removed or knocked out Math1, the mice did not develop the tumor, even when the sonic hedgehog pathway was activated.

"You need Math1 for that pathway to give you medulloblastoma," said Zoghbi. "This shows that there is a connection between normal development and cancer and underscores how important it is to study developmental biology. Cancer, in this case, is normal development gone awry."

One additional part of the equation is Gli2, a gene that is required for the signaling pathway of sonic hedgehog. Zoghbi and collaborators showed that Gli2 is regulated by Math1, thus explaining how Math1 regulates the proliferation of the cerebellar granule neuron precursors.

"If you take away Math1, there is no proliferation," said Zoghbi. "If you take away Math1, but add Gli2, proliferation continues."

"Math1 sets the cell up to respond to external stimuli," said Flora. If the cell cannot respond to the stimulation, it does not proliferate.

That raises the question of whether removing or inactivating Math1 might work as a treatment. That could depend on the stage of the brain's development, Flora said.

Most medulloblastomas arise after the age at which most of the cerebellar structure is complete. At that time, Math1 should be less active or even inactive in that structure.

"At least in theory, you could target these developmental pathways to at least slow down the growth of the tumor without impairing brain function," said Flora.