A researcher checks zebrafish tanks in a cancer lab. Credit: Stefano Dal Pozzolo/Contrasto/eyevine

Researchers have called for a common method of killing zebrafish used in laboratories to be abandoned amid growing evidence that it causes unnecessary suffering.

Experiment with distant quasar light could solve free-will loophole in quantum theory More humane method needed for euthanizing lab fish Microbial genes become useful forensic tool

The anaesthetic MS-222, which can be added to tanks to cause overdose, seems to distress the fish, two separate studies have shown. The studies’ authors propose that alternative anaesthetics or methods should be used instead.

“These two studies — carried out independently — use different methodologies to reach the same conclusion: zebrafish detect and avoid MS-222 in the water,” says Stewart Owen, a senior environmental scientist at AstraZeneca’s Brixham Environmental Laboratory in Brixham, UK, and a co-author of one of the studies. “As this is a clear aversive response, as a humane choice, one would no longer use this agent for routine zebrafish anaesthesia.”

The use of zebrafish (Danio rerio) in research has skyrocketed in recent years as scientists have sought alternatives to more controversial animal models, such as mammals. The fish are cheap and easy to keep, and although no firm data on numbers have been collected, millions are known to be housed in laboratories around the world. Nearly all will eventually be killed.

MS-222 (ethyl 3-aminobenzoate methane­sulphate, also known as TMS) is one of the agents most frequently used to kill the creatures. It is listed as an acceptable method of euthanasia by many institutions, and also by societies such as the American Veterinary Medical Association. But the study by Owen and his co-authors, published last year (G. D. Readman et al. PLoS ONE 8, e73773; 2013), and the second study, published earlier this month by Daniel Weary and his colleagues at the University of British Columbia in Vancouver, Canada (D. Wong et al. PLoS ONE 9, e88030; 2014), show that zebrafish seem to find the chemical distressing. The research should fundamentally change the practice, say the authors of both papers.

Owen’s study used video tracking to measure whether zebrafish avoided anaesthetics flowing through one side of a tank by moving to the other, untreated side. In the case of MS-222, the team found that zebrafish spent significantly more time in the untreated side than on the side containing the anaesthetic.

Weary’s team allowed zebrafish to first spend time in either a light or a dark section of a tank, and then exposed them to MS-222 on their preferred side, the light side. After exposure to the anaesthetic, all but one of 17 fish in the study spent less time on the light side, and nine completely avoided it. This indicates that the fish would rather undergo discomfort — in this case, darkness — than be exposed to MS-222.

“There must be something unpleasant” about MS-222 to produce such a strong signal in the experiment, says Weary, because fish do not avoid many other harmful chemicals to such an extent. “The results are pretty clear,” he adds. “We’re at a stage where it is a matter of getting policy-makers and researchers to think about this and to rethink the procedures.”

There is growing debate over the most humane methods of killing laboratory mammals, with rodent euthanasia coming under increased scrutiny (see Nature 500, 130–131; 2013). Fish euthanasia has so far attracted less attention. “I think of fish welfare as being 10 to 20 years behind mammal welfare,” says Lynne Sneddon, who studies welfare in fish and is director of bioveterinary science at the University of Liverpool, UK.

Sneddon says the two papers convincingly show that the use of MS-222 to kill zebrafish should probably be avoided. But she notes that there are significant differences between species — data on zebrafish should not be generalized to other laboratory fish, such as salmonids, for example — and therefore cautions against banning its use in the animals entirely.

Zoltan Varga, director of the Zebrafish International Resource Center at the University of Oregon in Eugene, also cautions against abandoning MS-222 because the optimal method of killing will depend heavily on the individual experiment and set-up. “A choice of anaesthetics is critical, as there is a range of reactions possible and we need to administer drugs that address any situation,” he says. In some cases, this could be MS-222.

There is not enough evidence to know which is the most humane method, and opinions differ. Owen suggests using the anaesthetic etomidate, which is cheaper than MS-222 (US$0.15 per litre compared with $0.23 per litre of working solution) and which seemed to be less aversive in his tests. Weary’s research suggests clove oil as another cheap alternative. Varga favours ‘hypothermal shock’ — in which the zebrafish, a tropical species, are rapidly cooled. (This method is illegal in the United Kingdom owing to concerns that ice may damage fish tissue while the animals are still conscious.)

As the number of fish experiments continues to rise — they are the second most popular research species in the United Kingdom — the question grows in importance. “We must have the patience to allow the zebrafish field research time to critically explore what the best — that is, most humane — standards are,” says Varga. “We can neither infer these standards from human experience nor from the guidelines and regulations already established for other laboratory organisms.”