Neurobiology

Edward A. Kravitz, Ph.D.

George Packer Berry Professor of Neurobiology

Kravitz Lab Website: http://www.hms.harvard.edu/bss/neuro/kravitz/

GENETIC MANIPULATIONS IN THE FRUIT FLY FIGHT CLUG: 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 steriod hormones serve important roles in the behavior. Our past studies have examined aggression using the American lobster Homarus americanus as a model system. Very recently, however, the major focus of our laboratory shifted to the examination of fighting behavior using common strains of the fruit fly, Drosophilia melanogaster. Although not widely known, 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 new experimental system for identifying the fundamental mechanisms underlying this behavior.

CURRENT RESEARCH
Fighting flies: genetic approaches to the study of their territorial behavior: Using a simplified experimental system we have been able to perform quantitative analyses of fighting behavior in male and female flies (see labworks.hms.harvard.edu). Towards this goal, we first assembled ethograms of the behavior (word descriptions of the components that make up fights), then videotaped and analyzed fights between pairs of male and pairs of female fruit flies, next constructed transition matrices and finally, used Markov Chain analyses to generate visual representation of average male and female fly fights (PNAS 2002 and 2004). With these in hand we have begun the following experiments.
1. Use of the GAL4/UAS system to alter amine neuron function in fly brains while flies are fighting: GAL4 is a yeast transcription factor that can be expressed in flies under the control of cell and region specific enhancer elements. The DNA target of GAL4 is a sequence of nucleotides called the Upstream Activation Sequence (UAS). In a second line of flies UAS constructs can be inserted into the genome driving the expression of any gene of interest. In the progeny of a cross between amine neuron specific GAL4 lines and UAS lines driving the expression of the shibirets gene (codes for the protein dynamin involved in synaptic vesicle recycling), synaptic transmission can be reversibly turned on and off by changing from the permissive to the restrictive temperature and back. Cell lines are available that will express GAL4 in (i) dopamine neurons and (ii) in dopamine and serotonin neurons, and we are attempting to generate GAL4 lines showing selective expression in (i) octopamine and (ii) serotonin neurons. Preliminary results show that turning off dopamine and serotonin neurons reduces the "willingness" of flies to fight.
2. Fighting in female flies and the role of genes of the sex-determination hierarchy in male- and female-specific patterns of aggression: Female fruit flies also show aggression and an examination of the patterns of the behavior in male and female flies reveals that many components of the behavioral pattern are shared but some are male-specific while others are female-specific. For example females show head to head interactions (head butting) that are not seen in males, while males show extended wing threats, flick their wings while retreating, show high intensity components of fighting like boxing, tussling and holding, and chase retreating opponents. In contrast to male fighting behavior, no clear hierarchical relationship results from the interactions between female flies. At present we are examining the role of genes of the sex determination hierarchy in determining male and female patterns of fighting behavior, and are using the transformer and fruitless genes to feminize regions of male fly brains at different times of development and in different parts of the nervous system to examine the consequences on the behavior.
3. Is there a long-term memory of social status in male fruit flies after winning or losing fights?: Recent studies show that male flies will fight for extended periods of time (up to 5 hours) and that the fighting strategy in winners and losers changes after a hierarchy is established. Flies remember familiar opponent after at least a 30 min separation period.
4. Other studies: We have begun preliminary studies examining patterns of gene expression in winning and losing flies in collaboration with Drs. Heinrich Reichert and Ronny Leemans in the Zoological Institute at the Biozentrum/Pharmazentrum in Basel, Switzerland and with Dr. Geoffrey Ganter at the University of New England in Maine. Collaborative studies also are underway with Dr. Mel Feany (Brigham and Women's Hospital and Harvard Medical School), who has generated fly models of human disease by expressing mutant human disease-related genes in flies. In these studies we are asking whether changes in aggression are seen in a Parkinson's Disease model fly line.