Most nerve cells have a single long axon and multiple short dendrites. The development of this polarity is crucial for the basic function of neurons. Writing in the Journal of Cell Biology, Michinori Toriyama and colleagues show that a newly discovered protein, shootin 1, has an important role in determining which branch from a developing neuron outgrows the others and becomes an axon.

Asymmetrical accumulation of shootin1 in the axonal growth cone. A polarized (stage 3) hippocampal neuron was double-stained with anti-shootin 1 antibody (red) and a volume marker CMFDA (green). Image courtesy of N. Inagaki, Nara Institute of Science and Technology, Ikoma, Japan.

In culture, hippocampal neurons extend several minor processes (neurites) during the first 12–24 hours after plating (stages 1 and 2); one of the processes then begins to elongate continuously to become an axon (stage 3). The transition from stage 2 to 3, which occurs 24–48 hours after plating, is the initial step of polarization. The researchers showed that the level of shootin 1 fluctuated in neuronal processes at stage 2. The higher its concentration, the faster the neurite was able to grow. This intracellular localization changed drastically at stage 3, with the protein predominantly accumulating in the nascent axon but disappearing from the sibling neurites.

The team found that the correct level of shootin 1 expression was necessary for establishing neuronal polarity. Overexpression of shootin 1 in hippocampal neurons led to its continuous accumulation in multiple neurites, resulting in the formation of surplus axons in nearly half the neurons. By contrast, repressing shootin 1 expression markedly delayed neuronal polarization, although most neurons were still able to become polarized by 7 days after plating.

But what regulates the asymmetrical distribution of shootin 1 in hippocampal neurons? Toriyama et al. noted that shootin 1 was actively transported from the cell body to neuronal processes as discrete boluses; this was accompanied by passive diffusion of the protein back to the soma. Inhibitors of actin or myosin abolished anterograde transport of shootin 1 and prevented its asymmetrical accumulation in neuronal processes.

As phosphatidylinositol-3-kinase (PI3K) is important for establishing neuronal polarity, the researchers wondered whether shootin 1 interacts with this pathway. They found that shootin 1 bound to PI3K in the rat brain, and that the proteins were also colocalized in axonal growth cones in stage 3 cultured hippocampal neurons. Importantly, abolishing shootin 1 expression reduced PI3K activity in axonal growth cones, whereas shootin 1 overexpression led to ectopic PI3K activation in multiple neurites, indicating that shootin 1 acts upstream of PI3K.

On the basis of these findings, the researchers propose a model in which anterograde transport and passive retrograde diffusion of shootin 1, which acts upstream of PI3K, provide a positive feedback loop that promotes neuronal polarization. So, if a neurite has more shootin 1 than its less fortunate siblings, it will outgrow them. This stretches the retrograde diffusion time of shootin 1, thereby increasing the protein level further and propelling even more neurite growth.