Fish smell danger

ANIMAL BEHAVIOUR Suresh Jesuthasan, a neuroscientist at the Biomedical Sciences Institutes in Singapore, and his colleagues isolated sugar molecules from the outer mucus of zebra fish. They found that when these molecules are broken into fragments, and added to water, they prompt alarm behaviour in other fish. This is a stepping stone for researchers to study neural circuits involved in fear responses, writes Amanda Schaffer

The hypothesis: Fish “smell” danger, and perhaps humans do, too. The investigators: Ajay Mathuru and Suresh Jesuthasan of the Biomedical Sciences Institutes in Singapore. When one fish is injured, others nearby may dart, freeze, huddle, swim to the bottom or leap from the water. The other fish know that their school mate has been harmed.

But how? In the 1930s, Karl von Frisch, the famous ethologist, noted this behaviour in minnows. He theorised that injured fish release a substance that is transmitted by smell and causes alarm. But von Frisch never identified the chemical composition of the signal. He just called it schreckstoff, or “scary stuff.”

Schreckstoff is a longstanding biological mystery, but now researchers may have solved a piece of it.

In a study published in Current Biology, Suresh Jesuthasan, a neuroscientist at the Biomedical Sciences Institutes in Singapore, and his colleagues isolated sugar molecules called chondroitins from the outer mucus of zebra fish.

They found that when these molecules are broken into fragments, as they might be when the fish’s skin is injured, and added to water, they prompt alarm behaviour in other fish.

At low concentrations, the fish were “mildly perturbed,” Jesuthasan said. At high concentrations, they stopped darting altogether and froze in place for an hour or longer. He and his colleagues also showed that neurons in the olfactory bulb of these fish were activated when exposed to the sugar fragments. In a sense, the fish seemed to “smell” the injury.

The work could have broad implications for understanding fear and panic in other animals, and perhaps in humans, said Lisa Stowers, a neuroscientist at the Scripps Research Institute who was not involved in the research.

Alarm signals

Researchers have long struggled for better ways to help patients who are chronically prone to panic or anxiety. Fear can be a useful tool for an individual animal. But it’s even more useful for one animal to be able to communicate its alarm – quickly – to others of its kind. Many lower animals seem to rely on smell to accomplish this, but surprisingly little is known of the substances used, or how they are produced or perceived. The best-known alarm signals are used by bees and ants.

The European honeybee releases a mixture of compounds after a sting. A major component is a molecule called isopentyl acetate, which rouses alarm in other honeybees.

“Carpenter ants release compounds called formic acid and n-undecane to signal danger to their fellows,” Jesuthasan said. “Ants that sense these chemicals stop moving, swing their antennae and then begin moving quickly.

If an enemy is spotted, they become aggressive. The exact response depends on the ratio of the chemicals.” Sea urchins release substances when their bodies are crushed that cause other sea urchins to flee.

Similar responses have been shown in marine snails, tunicates and tadpoles. But the chemical nature of the signals is not known, Jesuthasan added. In a 2008 paper published in Science, Marie-Christine Broillet, a neuroscientist at the University of Lausanne in Switzerland, identified the system responsible for picking up on alarm signals in mice:a few hundred neurons, called the Grueneberg ganglion, in the tip of the nose. But Broillet did not identify the signaling molecules – “a major scientific challenge,” she said. Chondroitin fragments may strike fear into the hearts of zebra fish, and perhaps even other fish, but they may mean nothing to other animals.


“Fear pheromones tend to be species-specific,” said Ajay Mathuru, a neuroscientist in Jesuthasan’s lab. Animals need to “send warning signals to their friends,” rather than “tip off the enemy,” he added, although examples do exist of predators picking up on the panicked cues of prey. Animals living in different environments may need different kinds of signaling molecules, said Marcus Stensmyr, an evolutionary biologist at the Max Planck Institute for Chemical Ecology in Germany.

Pheromones in humans?

Chondroitin fragments, which are relatively large, do not travel through air as they do through water, potentially limiting their usefulness to terrestrial animals. In humans, the notion that pheromones provoke fear and other responses remains controversial.

“Ninety-nine per cent of scientists probably don’t believe pheromones exist in humans,” said Murali Doraiswamy, a professor of psychiatry at Duke University Medical Center, although research “remains at a primitive stage.”

Certainly, visual cues, thoughts and memories play a more important role in generating fear behaviour in humans. “Even if you sprayed a person with a fearful perfume, if they could see with their eyes that nothing was happening, they wouldn’t react the way a fish would,” Doraiswamy said.

In zebra fish, the olfactory bulb has a close anatomical connection to a brain area called the habenula, which may also play a central role in mediating fear, Jesuthasan said.

In previous research, he and his colleagues disrupted signaling in the fish’s habenula and showed that this made them more likely to act “helpless” at the prospect of an electric shock. Those with abnormal signaling in the habenula had “an exaggerated fear response,” Jesuthasan said.

Role of the habenula

In humans, the role of the habenula is hotly debated. This is partly because the structure is small and situated deep in the brain, making it difficult to study using functional magnetic resonance imaging. It has been implicated in many different types of behavior, including stress, pain, anxiety, learning and reproduction, Stowers said. But for now, “its function is still a mystery.”

Even so, the brain areas implicated in the new zebra fish study could shed light on the neural circuitry involved in fear, even if those responses are set off by cues other than smell. And Jesuthasan’s group has purified molecules that can touch off alarm in the absence of other cues, an important step, Stowers said.

Now the researchers can “activate, mark and study the neural circuits involved in fear responses in a way that no one has before,” she said.