|
||||||||
|
NEUROBIOLOGY
Small Molecule Proves Its Punch Against Brain Tumor CellsNeurosurgeons must maintain a steely confidence during long hours in the operating room, but even they may surrender to moments of doubt. "I was at a meeting where a well-known surgeon specializing in brain tumors was saying he has this recurrent nightmare. He is operating on someone and he decides, 'This time, I'm really going to get all of the tumor,'" recounted Rosalind Segal, HMS associate professor of neurobiology at Dana-Farber Cancer Institute. "And in his dream he is operating and suddenly realizes he is removing the entire cortex."
AMD knocks out brain tumor cells. Without treatment, transplanted glial tumor cells grow uncontrollably in the skulls of mice (left). Treatment with AMD 3100 causes tumor cells to commit suicide, resulting in much smaller tumor masses (right). The tumors were stained with luciferase and viewed using the Xenogen IVIS in vivo imaging system. The spectrum of colors corresponds to luciferase intensity, with red being the most intense followed by yellow and blue. (Image courtesy of Andrew Kung) In real life, removing even a small brain tumor can rob patients of precious mental functions depending on where the cancer is located, making brain cancer one of the most difficult to treat surgically. Yet efforts to noninvasively and precisely eliminate brain tumors have met with little success. In the Nov. 11 Proceedings of the National Academy of Sciences, Rosalind Segal, Joshua Rubin, Andrew Kung, and their colleagues report that by aiming a small molecule at a receptor on the surface of human brain tumor cells, they were able to command the cells to commit suicide or, in some cases, to stop proliferating. What is more, the receptor, CXCR4, appears to be upregulated in many brain cancer cells, which in principle, means that blocking it with a small molecule could noninvasively destroy an entire tumor while sparing nearby healthy tissue. So far, the receptor-blocking drug, AMD 3100, appears to be remarkably effective. Brain tumors originate in either of two classes of cells, neurons or glia, and AMD 3100 targeted both types. It quelled cancer growth both in vitro and in human brain tumor cells transplanted into the crania of living animals. Progress Toward the ClinicFor clinicians and patients, the most exciting news may be that the tumor-thwarting molecule has already been shown to be safe and nontoxic in phase I clinical trials for a different disorder, HIV infection. "The translational part of this is very exciting," said Segal. Rubin, who is now at Washington University in St. Louis, is currently talking to several biotechnology companies and to clinical oncologists at Dana-Farber Cancer Institute and Brigham and Women's Hospital about launching clinical trials of CXCR4 antagonists with brain cancer patients. Some of these could be children since brain tumors are among the most common pediatric solid cancers, consisting of either neurons or glial cells. Adult tumors consist only of glial cells.
Her interest in CXCR4's possible role in brain cancer was piqued when DFCI researchers Scott Pomeroy, Todd Golub, and their colleagues showed, using gene chips, that CXCR4 is overexpressed in many brain tumors. Segal knew that CXCR4's ligand, CXCL12, is highly expressed by the new blood vessels surrounding brain tumors, but it was unclear what role the CXCR4-CXCL12 interaction was playing in brain cancer. To find out, Jennifer Chan, HMS research fellow in pathology at BWH, combed through brain tumor samples, both glial and neuronal, for signs of the pair. She found that CXCR4 was present at high levels in tumor cells and CXCL12 in endothelial cells, as measured by both RNA and protein expression. Rubin exposed cultured brain tumor cells of glial and neuronal origin to CXCL12. Not only did the cells express CXCR4, they did what CXCR4-activated immune cells do--they migrated, proliferated, and survived. Reversal of FortuneThe key step came when Rubin tried blocking the interaction with AMD 3100. "This small molecule antagonist had already been through clinical trials," Segal said. The trials had grown out of the discovery made in Springer's lab that CXCR4 is a coreceptor for HIV entry. Researchers scurried to find small molecules that might block the receptor. One of these, AMD 3100, proved safe in trials, though its efficacy against HIV infection remains unclear.
Rosalind Segal, Jennifer Chan, Andrew Kung (left to right), and their colleagues have found in AMD 3100 a brain tumor-thwarting molecule that may be both safe and effective. (Photo by Steve Gilbert) To Rubin's delight, AMD 3100 stopped the cultured brain tumor cells from migrating, proliferating, and surviving, even in the presence of the CXCL12 ligand. Working with Kung, an HMS assistant professor of pediatrics at DFCI, he introduced AMD 3100 systemically into mice whose crania had been transplanted with human brain cancer cells of glial and neuronal origin and observed the same tumor-defying effect (see figure on previous page). But it was not clear how AMD 3100 was affecting the cells. "Either fewer cells were being made or more were dying," said Segal. Staining the cells for signs of proliferation, on the one hand, and apoptosis, on the other, the researchers found a curious distinction. AMD 3100 inhibited proliferation and promoted apoptosis in the neuronal tumor cells, but seemed to affect only the apoptotic pathways in the glial cells. If AMD 3100 proves its mettle in a clinical trial for brain cancer, it could be joined by a fleet of CXCR4-blocking small molecules. "It is the only one to go through phase I clinical trials, but there are a large number of CXCR4 antagonists being developed," said Segal. Why bother with others if AMD 3100 works? Convenience is one rationale. In Segal's experiments, mice received AMD 3100 intravenously, either twice a day or continuously. "So the reason would be to find something that could be given once a day in a pill," she said. --Misia Landau |
