If you scare a fish, it will typically bend strongly to one side, making a C-shape, then kick away from you. A pair of neurons called Mauthner cells– one left, one right – can trigger the behavior. But a recent paper in the Journal of Neuroscience shows that Mauthner cells aren't the whole story.

Tsunehiko Kohashi from Nagoya University and Yoichi Oda from Osaka University used advanced microscopy techniques to describe the behavior of other cells, similar to the Mauthners, that also trigger escape behavior. The non-Mauthner escapes are a little slower to get started than normal Mauthner cell-mediated escapes, but once the behavior starts, the researchers couldn't tell the two apart.

Kohashi and Oda's results show that touching a fish's head tends to activate these other cells – called Mauthner homologs – more often than the Mauthner cells. The researchers also showed that the Mauthner homologs can modulate the strength of a Mauthner-mediated escape.

They used well-established optical techniques to measure the cell's electrical activity by injecting a fluorescent dye that changes fluorescence depending on the calcium levels in a cell. Calcium tends to increase when a cell fires an action potential, so monitoring the change in fluorescence let them see the activity in the neurons of interest. To get the dye into those neurons, it was linked to a sugar molecule that tends to be transported along axons to the cell body, and so the long axons of the Mauthner cells and their homologs – which extend all the way down the spinal cord –conveniently took up the dye and labeled the cell bodies in the hind brain.

The main innovation was to use a special high-speed confocal microscope that could rapidly scan between the different locations of the cells. Typically, Mauthner cells are located somewhat below the homologs, requiring rapid, precise control over the microscope's focus in order to get reliable measurements of changes in fluorescence in the two sets of cells simultaneously. To keep the fish stationary in the microscope to obtain a clear image, Kohashi and Oda embedded the fishes' heads in agar, leaving the tails free to flex during the C-start.

They were able to divide the behavior into two types, based on latency. Poking the ear or the tail produced a C-start with activity in both the Mauthner cells and the homologs about 4 ms later. Poking the nose, in contrast, produced an escape indistinguishable from the others, except that it took about 8–10 ms to get going and the Mauthner cells didn't spike.

Instead, these longer latency escapes were driven by the Mauthner homologs. Unlike the Mauthner cells, which fire a single action potential to set off the escape, the data showed that the homologs fire multiple times. And because their axons are much thinner than the Mauthner axons, the signal takes longer to travel down the spinal cord, which may be the reason why the movement starts a little later in non-Mauthner escapes.

Not only can the homologs trigger escapes on their own without the Mauthner cells but they can also work together with the Mauthner cells to modulate the strength of an escape. When a Mauthner cell and its homolog fire together, the researchers found that the resulting escape is stronger than when the Mauthner cell fires alone.

Other researchers had suspected that the homologs were involved in escape responses. Kohashi and Oda provided the first proof, and – more importantly – the first description of the neural activity underlying an escape response that doesn't require the Mauthner cells.

Kohashi, T. and Oda, Y. (
2008
). Initiation of Mauthner- or non-Mauthner-mediated fast escape evoked by different modes of sensory input.
J. Neurosci.
28
,
10641
-10653.