Sound and fury: Modulation of aggressive behavior through acoustic signals

In any sexually reproducing animals, one expects aggressive encounters to occur between/among males because of their adaptive value; competition over resources (food, territory, mates, social status) plays a critical role in both natural and sexual selection in most species of animals (1, 2). Male–male interaction, when behaviorally enacted, is labeled as aggression, and it varies in expression from subtle gestures to outright physically combative encounters. Moreover, aggressive behavior is usually mediated by species-specific signals in one or more sensory modalities.

In PNAS, Versteven et al. (3) continue the study of intermale acoustic signals in Drosophila melanogaster [e.g., Jonsson et al. (4)], advancing it by providing neurosensory mechanistic data by deploying classic lesion as well as genetic methods to probe the role of the fly’s “ears” in expressing aggressive behavior. The authors (3) show that the ability of hearing can affect the behavioral outcome of dyadic male–male encounters in laboratory contest arenas.

The auditory basis of hearing in D. melanogaster has been subject to numerous studies at multiple levels of biological organization over many size scales (5). The auditory hearing organ, known as Johnston’s organ (JO), is the modified second segment of the antenna. Previous work has shown that the JO is stimulated by sound-induced vibrations, such as airflow from the beating wings of a nearby fly. The salient acoustic signals for communication are produced by the male extending one wing and briefly fluttering it at the target receiver fly. This stereotyped gesture takes place with the sender and receiver at very close proximity, within one or at most two body lengths. The JO is easily visible and very accessible for experimental manipulation, which made deafening the fly relative easy using classic techniques that prevent acoustic vibrations from setting the JO into motion or simply ablating it (figure 1 in ref. 3).

Drosophila

Versteven et al. (3) investigated the consequences of deafening socially experienced D. melanogaster male flies by placing them in an experimental arena that normally promotes male–male aggressive interactions. Here, a deafened male was placed in an arena with an intact male and the interactions were video-recorded so that the number and level of aggressive encounters could be assessed. The assay proved to be sufficiently robust that reliable data could be obtained within 5 min of pairing. First, the investigators selectively extirpated one JO (unilateral aristectomy) and then both JOs (bilateral aristectomy). Compared with intact, control, males, the males that had the double aristectomy procedure showed reduced numbers of aggressive encounters. Males with bilateral aristal lesions exhibited slightly less aggressivity than the unilaterally lesioned males. A second, less drastic hearing-disruption procedure was tested by gluing (both unilaterally and bilaterally) the second and third antennae segments, thus drastically reducing their mechanical mobility, and presumably the sensitivity of the JO to acoustic stimulation. This procedure diminished aggressivity, compared with controls, but to a lesser extent than the aristectomy procedure. The most dramatic effect was on the duration of the aggressive interaction (“fighting time”). Whether the bilateral disruption to hearing was through aristectomy or gluing of the aristae, the male–male interaction time was diminished nearly fivefold compared with intact controls (5 s vs. 25 s). Although neither procedure entirely eliminated aggressive interaction, auditory/JO intactness had a pronounced effect on aggressivity. Auditory input clearly influences level and intensity of aggressive acts. This finding is not entirely surprising per se; however, Versteven et al. (3) pushed the analysis to the neurosensory level by applying genetic techniques that permitted them to genetically lesion neurons within the JO itself, while presumably sparing the intactness of the rest of the fly, including behavior.

Versteven et al. (3) performed several complementary genetic perturbations that strongly support the role of hearing in aggressive behavior and bolster the mechanical treatment results. First, the authors used a GAL4 line specific for AB neurons of the JO and GAL80ts to drive expression of tetanus toxin light chain, only at the adult stage. AB neurons transduce sound-evoked vibration. At the permissive temperature of 18 °C, tetanus toxin expression is blocked by GAL80ts and males in the aggression assay display normal levels of aggression. However, at the nonpermissive temperature of 25 °C, tetanus toxin is expressed, neurotransmitter release from the AB neurons is blocked, and aggression is significantly decreased. Another GAL4, also from the JO, but involved in gravity sensing—not hearing—does not have a significant effect on aggression.

Versteven et al. (3) initially examined genetic mutants that might alter aggression. From earlier work (6), a set of genes altered in hyperaggression mutants was compared with genes expressed in the auditory system and it was subsequently narrowed to those involved in signal transduction. The reduced list contained nan, iav, trpl, Arr2, inaD, and nompC (based on other criterion). All but the nan mutant altered aggression. However, nan and iav are broadly expressed and may have pleiotropic effects outside of the JO, and trpl, Arr2, and inaD are important players in the visual transduction, a modality known to be important in aggression.

To address the question of whether these mutant effects were specific, an experiment similar to the tetanus toxin study was performed, except this time, RNAi-mediated knockdown for each of the candidate genes was driven by the same JO AB-neuron GAL4 line in the presence of GAL80ts. In every instance, as in the RNAi experiment, males displayed reduced aggression if the genes iav, Arr2, nompC, nan, inaD, or trpl were knocked down. Thus, the specific RNAi knockdown matched the genetic mutant analysis in all cases except nan, firmly supporting the idea that sound-evoked transduction is an important determinant of aggressive behavior.

Rarely does a single sensory cue strictly control the release of a complex behavioral syndrome, such as social behavior, and in D. melanogaster it has long been known that in its mating behavior, courtship by males can involve cues from multiple sensory modalities, like vision, olfaction, taste, touch, in addition to hearing (7). It is reasonable to assume that aggressive behavior also engages multisensory cues. Versteven et al. (3) address such possibilities in their paper. The authors performed aggression assays using the “smell-blind” Orco mutant line and indeed, they report that mutant, smell-blind males exhibited a marked, 50% reduction in aggressivity compared with controls, in their assay conditions. Even so, however, performing bilateral aristectomy on Orco males further reduced aggressivity by another 30% or so, such that these doubly sense-disabled males were only about 20% as aggressive as control males. As in courtship behavior, the full behavioral syndrome of aggression is a multisensory affair.

Acoustic Aggression in Other Insects

Acoustic territorial and mating signals in the genus Drosophila have been well-documented in many species and, for example, the mating behavior in the drosophilid genus Zaprionus, where acoustic duetting behavior occurs and thus is even more complex than the analogous signals in D. melanogaster (7). This is worth noting because D. melanogaster is a model system in which modern tools of the genomics revolution are being applied to reveal the genetic control of the neural systems that underlie behavior, as indicated by the Versteven et al. (3) report. The decades-long hopes of making transgenic animals through gene transfer from species of “known genetics” to related species with homologous neural systems and behavior might even be on the horizon, as the CRISPR-Cas9 technology is developed. Two examples of interesting “non-melanogaster” drosophilids are presented to show interesting evolutionary variations on the theme of acoustic signaling in social behavior. Territorial songs and courtship duets are known in Drosophila virilis and the drosophilid Zaprionus genus (8). The second example comes from the most spectacular species radiation in Drosophila and is found in the Hawaiian Drosophila (9). It has been firmly established that territorial intermale aggression displays occur in the Drosophila-fauna of Hawaii, especially in the members of the Planitibia complex. Two members of this complex, D. heteroneura and D. silvestris, from the big island of Hawaii are thought to be lekking species (9). When males come into close proximity to each other they engage in conspicuous acoustic displays that can be recorded as substrate vibrational signals and such signals play a conspicuous role in their intersexual courtship, as demonstrated several decades ago (10). However, the auditory mechanism of their extensive signaling behavior, for either sound production or ears for hearing the sounds, is not yet known.

Studies of acoustic behavior in insects outside of Drosophila is extensive, particularly in crickets (11). Aggression songs accompany and punctuate the spectacular intermale cricket fights, where aggression escalates to a level of viciousness that the behavior was “elevated” to human spectator status and inevitably to the domain of blood sport gambling. The custom is ancient: cricket fighting tournaments are as old the Tang dynasty (approximately AD 600) (12). It has been reported that in cricket fights, the winner emits longer and more intense song bouts, driving the “loser” into relative silence. An interesting study was conducted by Phillips and Konishi (13), in which they discovered that when they deafened “loser-males” and allowed them to fight again—even against males that had previously defeated them—these deafened males paradoxically became more aggressive, including in their acoustic activity, even besting their former victors. The crickets were deafened by ablating their tympanal hearing organs, in a way comparable to the Versteven et al. (3) Drosophila study. Thus, the effect of deafening cricket males enhances, not diminishes, their combativeness. We raise these studies to remind investigators that what is true in the behavior of one species does not necessarily predict what will be obtained in other species.