The identification of neural subcircuits used by female fruit flies to make a choice about whether to copulate with a potential mate provides a template for understanding how the brain integrates complex information to reach decisions.
For humans, the decision to copulate is an intensely personal one, and we like to believe it is a choice made under free will. However, more than a century of studying other species has made it clear that specific internal states, together with the presence of particular external cues, can alter the probability of copulation in a consistent way across a population, strongly suggesting that there are neural circuits that evaluate relevant, predetermined variables and so bias behaviour. Decision-making circuits are present in all species with a nervous system, and understanding how the brain carries out this type of computation is a major goal of neuroscience. Now, three studies (one published in the journal Current Biology1 and two in Neuron2,3) using different genetic strategies have identified circuit components that control the receptivity of female fruit flies to male courtship, outlining the scope of this complex decision-making process.
Sexual behaviour in the fruit fly Drosophila melanogaster is a particularly useful model system for studying decision-making because it involves both stereotyped and plastic features. The complicated sex-related behaviour of flies is conducted by a relatively small brain, and investigators have an arsenal of sophisticated genetic tools with which to reproducibly identify and manipulate particular neurons in freely behaving animals4. Although the courtship behaviour of males has been studied intensely, progress in understanding female reproductive behaviour5,6 has made it apparent that females are not simply passive recipients of male advances. Instead, female flies engage in an active and complex decision-making process5,6 that determines whether copulation occurs. The female's decision-making apparatus uses sensory information — including courtship songs produced by male wing vibration, visual cues and olfactory cues such as pheromones — to assess male fitness in the context of the internal state of the female herself (Fig. 1).
Bussell et al.1 undertook a genome-wide screen to look for genes that alter the receptivity of female flies to potential mates. They found that decreasing the activity of the transcription factor Abdominal-B (Abd-B) in female neurons decreased the rate of mating. The authors showed that neurons expressing Abd-B during development regulate the rate of female pausing during courtship, an indicator of receptivity. Abd-B-expressing neurons were activated in response to male-specific sensory inputs such as courtship song, but only in the presence of male flies (playback of a recording of the song alone was ineffective, indicating that the male probably provides additional visual or chemosensory cues). The fact that these neurons are downstream of sensory inputs suggests that Abd-B neurons are part of the receptivity-output arm of the fly decision-making circuitry, and are driven by higher centres that process and integrate courtship-relevant information.
In their hunt for circuits involved in this sensory integration, Zhou et al.2 began with the assumption that neurons that express doublesex (dsx), a gene that is differentially expressed in male and female reproductive circuits, would contribute to female-specific behaviour6. The authors used state-of-the-art genetic techniques4 to identify and manipulate small populations of neurons expressing Dsx protein in the female adult brain, and found that activation of two neuron groups, pC1 and pCd, enhanced the rate of copulation.
Zhou and colleagues provide anatomical evidence to suggest that pC1 and pCd convey information between brain areas known to be involved in processing courtship-related sensory information5,6. Using calcium levels as an indicator of neuronal activation, the authors showed that pCd was responsive to cis-vaccenyl acetate, a male-specific lipid pheromone that enhances female receptivity. Male song activated pC1, and this response was enhanced by the presence of cis-vaccenyl acetate, suggesting that pC1 neurons convey integrated information. Thus, pC1 and pCd are part of the central circuitry that processes courtship-related information.
In addition to extrinsic cues, receptivity is dependent on the female's internal state. Females that have recently mated do not copulate, even if presented with a fit and eager male. This change in behaviour is due to transfer of the protein sex peptide (SP) to the female in the male's seminal fluid. SP-responsive receptors have been identified7 on nerve cells in the abdominal ganglion (a neuronal structure roughly analogous to the spinal cord) that innervate the reproductive tract, but it was unknown how the SP signal is transmitted to the central nervous system.
Feng et al.3 looked for groups of neurons that decreased female receptivity when electrically silenced. Their screen identified a group of neurons that project from the abdominal ganglion into the brain, which they named SP abdominal ganglion (SAG) neurons. The authors observed that these neurons are not themselves responsive to SP, but rather receive information from SP-sensitive neurons through synapses (junctions that transfer signals between cells). Crucially, the strength of this connection was modulated by SP and correlated with the female's mating status. These data establish SAG neurons as a conduit of information on mating status from the reproductive tract to the central nervous system.
Except to those of us who are dedicated fly voyeurs, the importance of these individual studies might be debatable. In aggregate, however, they provide a framework that could lead to a detailed cellular and molecular understanding of a multifactorial decision-making process. By highlighting three different and as-yet-unconnected regions of the female fly's sexual-behaviour circuitry, these studies provide starting points for completing the wiring diagram.
Flies are faced with many of the same basic challenges as humans: what to eat, when to sleep and whom to mate with. Choice — our exercise of free will — is the probabilistic representation of integration processes that are rooted in the molecular and neural architecture of our brains. The genetic and electrophysiological tools available in the fly make this model organism arguably the best place to get our first glimpse into how a brain can make complex decisions.
About this article
Zoological Science (2016)