The courtship rituals of fruitflies are disrupted by mutations in the fruitless gene. A close look at the gene's products — some of which are sex-specific — hints at the neural basis of the flies' behaviour.
Richard Feynman is reported to have said, “Science is a lot like sex. Sometimes something useful comes of it, but that's not the reason we're doing it.” In three papers, two published in Cell1,2, and one in this issue3 (page 395), science and sex have come together to provide us with something useful — an extraordinary glimpse into how the male and female nervous systems function to generate sexual behaviour in fruitflies (Drosophila).
Unlike many British males on a Friday night, Drosophila males do not simply jump on the first female they see. Courtship behaviour in D. melanogaster is a stereotyped and instinctive sequence of behaviours performed by the male, involving visual, olfactory, gustatory, tactile, acoustic and mechanosensory stimuli being exchanged between the sexes (Fig. 1). The female's role is considerably less dramatic than the male's: she simply runs away, gives the odd kick, then mates (or not)4.
Normal mature males seldom court other males, but male fruitless mutants are bisexual, courting not only females but also other males5. In exclusive male company, these mutants can form bizarre courtship chains, where several males, each chasing and courting the one in front, generate frenzied revolving circles.
The gene mutated in the fruitless flies (termed fru) was molecularly cloned in 1996, and the putative protein that it encodes was identified as a transcription factor6,7, a regulatory molecule that controls gene expression. A large number of different messenger RNAs can be generated from the fru gene, some of which are sex-specific. In particular, an mRNA produced only in males is translated into a protein called FruM (for male-specific Fruitless)6,7. This sex-specific production of the fru mRNAs is determined by the canonical sex-determination system, the most relevant component of which is the transformer gene (or tra).
Briefly, the encoded Tra protein binds to very short sequences (13 nucleotides) on the immature fru mRNA, to sex-specifically regulate which portions will be ‘spliced’ into the final transcript6,7. (Indeed, Ryner et al.6 cloned fru by looking for genes that contained Tra-binding sequences.) Similarly, Tra protein binds to the doublesex (dsx) gene and splices it in male- and female-specific modes (DsxM and DsxF, respectively)8. The DsxM and DsxF transcription factors mainly determine sexual morphologies8, but the sexual identity of the nervous system is shaped by fru.
By forcing males to express the female-specific fruF transcript, Demir and Dickson1 produced males that showed the characteristics of the worst-affected fru mutants. These males were sterile, they barely courted females and they were more interested in courting males, forming courtship chains. By contrast, females jammed into fruM mode mated poorly, produced very few eggs, but — astonishingly — courted other females (Fig. 2), even to the point of forming chains. And an identity crisis of similar epic proportions was observed in females that were ‘masculinized’ using a different fru-related genetic trick3. Finally, by feminizing specific abdominal glands in males to produce female pheromones, and placing the altered males with fruM females, the sex roles were reversed, so that the females courted the males1.
In another nifty piece of genetic engineering, both teams2,3 generated flies in which they could, among other things, mark the parts of the nervous system (just 2%) that show sex-specific expression of Fru. Further genetic manipulations showed that high levels of male–male courtship result when the communication between these neurons is shut down, or when fruM expression in these neurons in males is inhibited2,3. Both studies found that the central nervous system of males and females looked very similar in terms of sex-specific fru expression, with few differences between the sexes in the numbers, positions or wiring of cells expressing Fru.
The fru products were found in almost all sensory organs that have been implicated in courtship2,3. Olfactory sensory neurons showed some evidence for sexual dimorphisms. Those receptors that respond to pheromones project to certain other brain regions that are larger in males than females, reflecting the fact that sex pheromones have a greater functional significance in male Drosophila2. By reversibly shutting down the fru-expressing olfactory receptors, both in males and in masculinized females in the sex-reversal paradigm outlined above, court-ship behaviour declined significantly, implying that these receptors are central to sexual interactions2. However, by decreasing FruM in males just in these neurons, homosexual courtship increased, so normally these olfactory receptors must inhibit male–male interactions3.
So, a single fru-encoded genetic switch seems to be sufficient to shift the functioning of the nervous system from male to female mode, irrespective of the morphological sex of the animal. The general absence of large-scale sexual dimorphisms in fru-expressing neurons implies that it is the molecules regulated by fru that make the difference. Future work will undoubtedly be aimed at finding these molecules, as well as identifying the subset of key neurons that are sufficient to gener-ate male courtship elements. Indeed, Villela et al.9 have identified neurons downstream of ones expressing fru that are implicated in the control of the male's courtship song. Finally, an intriguing and mostly forgotten paper was published 30 years ago10 about ‘lesbian’ Drosophila females that courted other females — apparently because of a genetic factor(s) on chromosome 2 (fru is on chromosome 3). Might this long-lost strain have carried a mutation in one of the fru target genes?
The work discussed here may well find itself the focus of attention for those interested in the debate (scientific and political) on the genetic versus environmental bases of human sexuality. Perhaps we should remind ourselves that normal fly sexual preferences, unlike human sexual behaviour, cannot be modulated to any significant extent by altering experience11.
Demir, E. & Dickson, B. J. Cell 125, 785–794 (2005).
Stockinger, P., Kvitsiani, D., Rotkopf, S., Tirián, L. & Dickson, B. J. Cell 125, 795–807 (2005).
Manoli, D. S. et al. Nature 436, 395–400 (2005).
Greenspan, R. J. & Ferveur, J. -F. Annu. Rev. Genet. 34, 205–232 (2000).
Villela, A. et al. Genetics 147, 1107–1130 (1997).
Ryner, L. C. et al. Cell 87, 1079–1089 (1996).
Ito, H. et al. Proc. Natl Acad. Sci. USA 93, 9687–9692 (1996).
Cline, T. W. & Meyer, B. J. Annu. Rev. Genet. 30, 637–702 (1996).
Villela, A., Ferri, S. L., Krystal, J. D. & Hall, J. C. Proc. Natl Acad. Sci. USA (in the press).
Cook, R. Nature 254, 241–242 (1975).
Siegel, R. W., Hall, J. C., Gailey, D. A. & Kyriacou, C. P. Behav. Genet. 14, 383–410 (1984).
About this article
Behavioral testing regimens in genetic-based animal models of Parkinson's disease: Cogencies and caveats
Neuroscience & Biobehavioral Reviews (2013)
Parkinson's Disease (2012)
The Journal of Sexual Medicine (2009)
Modulatory Communication Signal Performance Is Associated with a Distinct Neurogenomic State in Honey Bees
PLoS ONE (2009)