Two research groups in Asia have identified an enzyme that governs whether a neural projection becomes an axon or a dendrite and several of its upstream and downstream targets. Their findings on the function glycogen synthase kinase-3β (GSK-3β) were reported in the January 14 issue of
In the first paper, scientists from the Shanghai Institutes of Biological Sciences and the National Institute of Biological Sciences in Beijing, led by Yi Rao, reported that GSK-3β activity is inversely related to axon growth.
Rao, now the associate director of Northwestern University's Institute for Neuroscience, Chicago, told
To test the enzyme's role in establishing neural polarity, the researchers created constitutively active GSK-3β mutant rat hippocampal neurons, which formed fewer axons than control neurons but showed no difference in dendrite numbers. Inhibiting GSK-3β in normal hippocampal cells resulted in the formation of multiple axons and fewer dendrites.
"[This] provides consistent results that the level of GSK-3 activity is critical," said Rao. His group also identified two proteins upstream of GSK-3β that affect its activity: PTEN, a lipid and protein phosphatase, regulates Atk, an axon-localized kinase, which in turn regulates GSK-3β.
The second team, from Nagoya University and Hiroshima University in Japan, identified a downstream target of GSK-3β. They reported that GSK-3β controls the formation of axons and dendrites by regulating the activity of collapsing response mediator protein-2 (CRMP-2), which is responsible for assembling the microtubules that enable neural projections to grow. By treating hippocampal neurons with neurotrophin-3, the team inactivated GSK-3β, which decreased CRMP-2 phosphorylation and promoted axonal growth.
Lead researcher Kozo Kaibuchi, a professor of cell pharmacology at Nagoya University, explained, "It has been reported that CRMP-2 is hyperphosphorylated in the soluble fraction of the brain in Alzheimer disease. So, we tried to find the responsible kinase and identified it as GSK-3."
Mineko Kengaku, head of the laboratory for neural cell polarity at Japan's RIKEN research institute, who was not involved with either of the studies, said: "GSK-3β is a multifunctional protein that regulates activities of several microtubule-associated proteins, transcription factors, Wnt signals, and so on."
Rao added that establishing polarity is also important for other types of cells, including gut epithelial cells. Some of the same proteins might be involved, he said, although that remains to be tested.
Kaibuchi agreed: "I assume that the upstream signaling from GSK-3 may be common among other polarized cells, including fibroblasts. Since CRMP-2 is abundant in neurons but not other tissues, the substrate of GSK-3 may be different in other cells."
GSK-3 is also essential for maintaining cell polarity, Rao said. "The polarity, once formed, needs an active mechanism to be maintained, otherwise you go back to a non-polarized neuron," he explained.
Rao's team converted pre-existing dendrites into axons by inhibiting GSK-3 activity, and he envisions a clinical application for this research: "Can one, in the future, convert functionally redundant dendrites into axons? …If that's possible, then there's a new way to make axons after [nerve] truncation."
Kaibuchi agreed. He said that in a 2003 study, his group demonstrated that CRMP-2 overexpression can enhance regrowth of damaged axons in rats. "Thus, GSK-3 becomes a nice therapeutic target for regrowth of axons after nerve damage and potentially for Alzheimer disease."
Kengaku said this approach might be more successful than using stem cells. "Intensive efforts on neural stem cells have raised new possibilities for repairing the nervous system," she wrote. "To reconstitute the neural network, however, one should put the neuron back at the right place and let it form the dendrites and axons in correct shape and orientation."