Figure 3. Aggressive behavior is disrupted by courtship patterns in males without Fru^M function in OA neurons (A–D).

Four separate male pairings in fight chambers were examined. Two control males, UAS-fru^MIR/+ (white columns, n = 10), two males without Fru^M function in OA neurons, dTdc2-Gal4/UAS-fru^MIR (black columns, n = 17), one control male UAS-fru^MIR/+ and one dTdc2-Gal4/UAS-fru^MIR male in which the control male was the dominant male, (light grey column, n = 7), and one control male UAS-fru^MIR/+ and one dTdc2-Gal4/UAS-fru^MIR male in which the dTdc2-Gal4/UAS-fru^MIR male was the dominant male, (dark grey column, n = 5). (A) Differences in the latency to first encounter between groups were not observed, (F(3,35) = 0.19, P = 0.9, ANOVA for independent groups). (B) Male pairs engaged in similar numbers of encounters in a 30 min fight, (F(3,35 = 1.59, P = 0.2, ANOVA for independent groups). (C) The winning male in each pair performed a similar number of lunges per 30 min fight, (F(3,35) = 0.4, P = 0.75, ANOVA for independent groups). (D) Males that lack Fru^M function in OA neurons, dTdc2-Gal4/UAS-fru^MIR males, displayed unilateral wing extensions followed by abdomen bending and attempted copulation whether paired together (black column), or as the dominant or subordinate male when paired with a control male (light and dark grey columns). (A negative binomial model (with log link function) was utilized. A likelihood ratio test of the equality of the group means yielded a highly statistically significant result (P<0.00006). Comparisons of groups yielded statistically significant differences between dTdc2-Gal4/UAS-fru^MIR males male pairings and control UAS-fru^MIR/+ males (Bonferroni-adjusted P = 0.0024) and between the dark gray column (one UAS-fru^MIR/+male and one dTdc2-Gal4/UAS-fru^MIR male with the dTdc2-Gal4/UAS-fru^MIR fly establishing dominancy) and control UAS-fru^MIR/+males (white column) (P = 0.0091)).