Make mine a double

When Devineni isn’t watching fly mutants with names like cheap date, tipsy and bar fly staggering around in jars, she’s asking questions about the choices her flies make. “We had no idea whether flies would want to drink alcohol or not,” she says. Heberlein’s lab has been studying flies and alcohol for several years, and has identified a number of genes involved in alcohol tolerance and addiction, later validated in rodent models. But when it comes to complex behavior such as choosing to drink, it’s been a little bit more difficult to do the experiments.

Previous work was limited to using a device called an “inebriometer,” which emits gaseous ethanol. But when flies have no choice but to breathe in ethanol, how do you measure if they are choosing to consume it?

The turning point came just as Devineni started the project in 2007, when a paper appeared describing a new feeding assay (Proc Natl Acad Sci USA 104:8253–56, 2007). Called the “Capillary Feeder,” or CAFE for short, it’s little more than a jar with capillary tubes inserted into the lid. The tubes are filled with liquid fly food—a mix of yeast extract and sugar. As flies in the jar drink from the tubes, the level of food drops. Devineni says she can measure when the flies drink as little as 50 nanoliters. “We can directly see every time the flies take a drink.”

Armed with the CAFE assay, Devineni let flies decide what to drink, and how much. She found the results surprising. “I said, ‘Wow! Did you guys really drink that much?’” Her flies preferred alcohol in their food, and chose to increase the amount of ethanol they drank until they began to lose control of their legs. At this stage, flies contain about 35 millimolar ethanol, the human equivalent to twice the legal driving limit in the United States. But then, says Devineni, they stopped; flies do not binge drink.

Even after removing flies’ ability to smell alcohol, and spiking ethanol-containing food with bitter quinine, the flies still favored the alcohol food. “Their drive to drink is pretty robust,” she says. “I was surprised how strong the attraction was.” When the flies were forced to go cold turkey for a couple of days (a long time in the lifetime of a fruit fly) they rapidly returned to their old ways and went straight for the ethanol-containing food (Curr Biol, 19:126–32, 2009).

So flies have a mix of things going on: taste repulsion, olfactory attraction, and feeling intoxicated. Devineni says, “Alcohol drinking is a pretty complex behavior that is a balance of these different factors.”

Rainer Spanagel, who studies the genetics behind drug abuse at the Central Institute of Mental Health in Germany, helped select this paper as one of Faculty of 1000’s Hidden Jewels. “I cannot think of any further control they could do,” Spanagel says. “It’s a really brilliant paper.”

The idea that fruit flies can model a complex behavior such as addiction was initially very controversial, according to Devineni. Alcoholism itself is a human condition influenced by social and cultural as well as genetic and physiological factors. Although she would be the last to claim that addiction in flies and humans is identical, Devineni is confident that they now have laid a good foundation for identifying molecular mechanisms of addiction and dependence. Spanagel agrees. “If a human genetic screen gives a hit, we can make a fly mutant, do the CAFE assay, and look for a change. If it matches the human data, we’ve validated it.”

The Hidden Jewel describes a recent paper from a less obvious journal, selected by the Faculty. Click here to see the full F1000 review.