Substrate-borne vibratory communication during courtship in Drosophila melanogaster

Abstract

Courtship in Drosophila melanogaster has become an iconic example of an innate and interactive series of behaviors. The female signals her acceptance of copulation by becoming immobile in response to a male's display of stereotyped actions. The male and female communicate via vision, air-borne sounds, and pheromones, but what triggers the female's immobility is undetermined. Here, we describe an overlooked and important component of Drosophila courtship. Video recordings and laser vibrometry show that the male abdomen shakes ("quivers"), generating substrate-borne vibrations at about six pulses per second. We present evidence that the female becomes receptive and stops walking because she senses these vibrations, rather than as a response to air-borne songs produced by the male fluttering the wings. We also present evidence that the neural circuits expressing the sex-determination genes fruitless and doublesex drive quivering behavior. These abdominal quivers and associated vibrations, as well as their effect on female receptivity, are conserved in other Drosophila species. Substrate-borne vibrations are an ancient form of communication that is widespread in animals. Our findings in Drosophila open a door to study the neuromuscular circuitry responsible for these signals and the sensory systems needed for their reception.

Figure 1

Two Behaviors of the Male Vary in Frequency with Respect to whether the Female Is Moving or Immobile (A and B) Frequencies were extracted from the ethograms built from movies of courting pairs. In (A), the x axis shows the values for each of 30 pairs of Oregon-R flies. The y axis shows the percentage of the time the males display wing fluttering (including wing extension/vibration and scissoring) or abdominal quivering. The left scatterplot shows these two male behaviors when the females are moving, and the right one the male behaviors when the females are stationary. Note that one male behavior is shown without indicating whether the same male is also performing the other behavior at that time. Therefore, the table in (B) breaks down male behavior further, showing for each behavior the grand means (n = 30, as before) as percentage of the time the female is moving or immobile. See also Figures S1A and S1B. All intervals in this report are given for a 95% confidence level. (C) Log-linear models of association were tested (see Figure S1D), and the best fit includes a strong association between male quivering (or not) and female movement (or not) and a weaker but still significant association between male fluttering (or not) and female movement (or not). See also Figures S1 and S2, Movie S1, and Movie S2.

Figure 2

Behaviors of Neuronally Masculinized Females during Courtship with Wild-Type Females Masculinized females, like wild-type males, quiver their abdomens and the wild-type female partners appear to respond by stopping. Data are presented as in Figure 1. See also Figure S3 and Movie S4.

Figure 3

Substrate-Borne Vibrations Generated during Abdominal Quivering of Courting Males (A) Scheme of the video and laser vibrometer recording system. (B) Oscillogram of substrate-borne vibrations generated during a single bout of quivering of about 7 s; the wings of the male were amputated. There is some variation in the amplitudes of the substrate vibrations. (C) Details of (B) above to show higher resolution. See also Movie S5.

Figure 4

Behaviors of Courting Drosophila yakuba and D. sechellia Pairs An association between male quivering and female immobility is apparent. (A and B) Data are presented as in Figure 1. (C) Oscillograms of substrate-borne vibrations generated during a single bout of quivering show that the vibrations are similar to D. melanogaster. See also Figure S4 and Movie S6.