Edward Kravitz

  • Kravitz Lab

GENETIC MANIPULATIONS IN THE FRUIT FLY FIGHT CLUB: Aggression is a nearly universal feature of the behavior of social animals. In the wild, it is used to procure food and shelter, for protection from predation and for selection of mates, all of which are essential for survival. Despite its importance, little is known of the neural mechanisms that underlie aggressive behavior, other than that hormonal substances, including amines, peptides and steroid hormones serve important roles in the behavior. Our past studies have examined aggression using the American lobster Homarus americanus as a model system. About a dozen years ago, however, the major focus of the laboratory shifted to the examination of fighting behavior using common strains of the fruit fly, Drosophilia melanogaster. Although not widely known at the time, fruit flies do fight and males at least become territorial (establish dominance relationships). With the genome fully sequenced and with elegant methods available for the selective manipulation of genes in subsets of central nervous system neurons, behavioral studies of aggression in flies offer a powerful experimental system for identifying the fundamental mechanisms underlying this behavior.

CURRENT RESEARCH: Early results with this model system demonstrated that male and female flies compete over resources in same sex pairings. Male fights lead to the establishment of hierarchical relationships while female fights end up with flies sharing resources. In male fights, defeated flies develop a “loser mentality” in which they will lose all subsequent fights with either familiar or unfamiliar opponents, although they fight differently against opponents in the two cases. The duration of the memory of losing is dependent on the training protocol, with a “spaced” training protocol leading to the “loser mentality” being extended for up to a week. At present there are two main themes under investigation in the laboratory. I.The role of amines in aggression: how do amine neurons work? In earlier studies when we manipulated entire pools of serotonin, dopamine or octopamine (fly equivalent of norepinephrine) neurons (100-200 of each subtype) in aggression or other social behaviors, we generated interesting phenotypes. For example, we found that serotonin was not required to initiate aggression, but was involved in facilitating the transition to higher-level aggression during fights. By contrast octopamine, appeared to be involved in the choice between courtship and aggression and not involved directly in transitions to different intensity levels. Dopamine deficiency generated hyperactive flies that did not interact in social behaviors with other flies. In more recent studies aimed at better understanding of the roles of amines, we generated a method that allowed us to reproducibly look at amine neurons one by one. To perform these experiments, we developed an intersectional genetic method that allows us to restrict the numbers of neurons isolated of each subtype by combining the binary Gal4/UAS system with the flp/frt recombinase technique. Using this method, we identified a single pair of serotonin neurons that appear to be involved in going to higher intensity levels in fights, and two pairs of dopamine neurons that appear to do the opposite: they limit the ability to go to higher intensity levels. These pairs of single neuron types are specialists in aggression and are not involved in other social behaviors of fruit flies. Thus via this approach we believe we have identified key control points in the neuronal circuitry associated with aggression in flies and have generated tools and methods that allow us to manipulate these control points in live, awake and socially interacting flies. Through a different novel genetic technique we can identify the pre- and post- synaptic contacts of the amine neurons, and thereby begin to unravel parts of the circuitry that governs a complex behavior like aggression. II. The “loser mentality” and other long term aggression-related changes in behavior induced during development or through fighting experience as adults: The experience of winning or losing in male fly fights can lead to long term changes in the behavior of flies in subsequent fights. The “loser mentality” described above is being explored in the laboratory at the present time as a potential model of “chronic” or “conditioned” defeat by asking if other behaviors like sleep, feeding or courtship success also are influenced in loser flies. Drugs that are used in clinics to treat various psychiatric conditions in humans also are being examined to ask whether they influence any of the parameters altered by chronic defeat. Inbreeding of winners in fights can lead to the generation of “bullies”, who will win all fights against the Canton-S starting parent line. The “bullies” begin fights earlier, accelerate to higher intensity levels more quickly and always retaliate against “lunging” by an opponent, in their fights. In contests between paired bullies, however, a loser is generated who loses all competitive advantage against all opponents for a short period of time. They revert to bullies again, however, over the next several hours. We are very interested in the nature of the changes that take place in the nervous systems of male flies to create the hyper-aggressive “bully” phenotype. Whatever the nature of those changes, they occur during a short window in development during the pupal life of flies.

This material is based upon work supported by the National Science Foundation under Grant No. 0751650 and by the National Institutes of General Medical Sciences under grant Nos. GM067645 and GM074675.
Any opinions, findings and conclusions or recommendations expressed in this material are those of the the Kravitz Laboratory at Harvard Medical School and do not necessarily reflect the views of the National Science Foundation (NSF) or the National Institutes of General Medical Sciences (NIGMS).

We are very interested in the nature of the changes that take place in the nervous systems of male flies to create the hyper-aggressive 'bully' phenotype. Whatever the nature of those changes, they occur during a short window in development during the pupal life of flies.