Finding a new behavioral adaptation in fruit flies

In the game of evolution, key behavioral adaptations that confer fitness in survival and reproduction, paying tremendous dividends for an individual’s progeny, may seemingly arise from thin air—so much so, even familiar species like the humble fruit fly can surprise biologists.

In a paper in Nature Communications, Yun Ding’s lab in the University of Pennsylvania’s School of Arts & Sciences discovered a novel courtship behavior Ding calls “wing spreading” in female Drosophila santomea, a fruit fly species native to the island of São Tomé off the coast of West Africa.

“When it comes to courtship,” Ding says, “most studies of fruit fly species focused on how male behaviors such as courtship songs lead the interaction, with females playing a more passive role.” But the team’s work on D. santomea found that the females take the lead, signaling their receptivity by spreading their wings, actively engaging with the male song, and driving the courtship rhythm.

In with the new

The discovery of this newly evolved novel behavior is particularly exciting to Ding’s evolutionary neurobiology lab because they now have a system to understand where behavioral novelty comes from in the nervous system.

“Many species-specific traits are often ancient behaviors resurfacing under the right conditions,” Ding says. “At first, we were cautious in our claim of novelty, but our methods and analysis continually pointed to this female wing-spreading behavior being a newly evolved novelty.”

Such novelty, Ding says, is the rare kind that scientists in evolutionary biology live to uncover: a change that can arise through minute adjustments in neural circuits, creating what appears to be an entirely new behavior within a species.

To understand what encodes the novel wing spreading behavior, Dawn Chen, co-first author and postdoctoral researcher in the Ding Lab, says the team undertook a blend of comparative behavioral and neural techniques, spanning both physical manipulation and genetic approaches.

“We first mapped the neural circuits of D. santomea females to identify which neurons triggered wing spreading in response to the male song,” Chen says. “Then, we used optogenetics, a technique that uses light to activate specific neurons, and applied it across D. santomea as well as its close relatives, D. yakuba and the popular model species D. melanogaster.”

Eventually, the scientists found that evolution has reused a pair of neurons that regulates a conserved female abdomen behavior to also control wing spreading in D. santomea females.

Turning up the heat

The paper’s other co-first author, Minhao Li, describes how this discovery was bolstered by observing D. melanogaster, which do not normally perform wing spreading, in a different condition. “When we raised D. melanogaster in higher temperatures,” Li says, “we observed that females showed a latent wing-spreading response.”

This dormant ability in D. melanogaster hinted that, even in species without the behavior, there are underlying neural circuits that may facilitate it. “This was the most exciting moment of our study,” Li says, “because it revealed that these circuits were there all along, waiting for the right trigger to unlock them.”

Chen points to the implications of these findings in understanding how neural circuits drive social behaviors. “The fact that a subtle change of neurons can spark new, socially interactive behaviors like wing spreading suggests that evolution might lean on existing circuits to develop new forms of communication.”

The researchers used a 96-channel recording apparatus Ding and collaborator Steven Sawtelle developed called the "SongTorrent." It's a likened to a film set, simultaneously capturing audio and video of the flies' courtship bouts. 

This phenomenon, where existing pathways or circuits adapt to serve new functions, is known as neural co-option and may offer a powerful lens for understanding how behaviors evolve, Ding suggests. “There’s more to nervous systems under the surface,” she says. “They have remarkable potentials to encode new behaviors, and sometimes these potentials can be actualized during evolution to produce novel behaviors.”

Looking ahead

For the team, the next steps involve diving deeper into how environmental factors might influence these latent circuits. “The temperature experiments were an eye-opener,” Ding says. “Could it be that other latent behaviors are hiding in plain sight, ready to emerge when the environment shifts?”

Looking ahead, the team hopes to use these findings to advance the field’s understanding of how neural circuits evolve. By tracing these neural circuits in greater detail, Ding and her colleagues aim to piece together a more comprehensive view of the neural basis for courtship behaviors. “Ultimately, we want to see how these latent potentials could represent a broader theme in evolution,” Ding says. “Our study suggests that behaviors are not always de novo creations but are often waiting to be unlocked in just the right way.”

This work, the researchers believe, also sets a foundation for exploring the often-overlooked role of females in courtship evolution. “Traditionally, male behaviors have been the center of the spotlight in courtship behavior research,” Li says. “But with D. santomea, we see that female behaviors can drive dynamic social interactions in ways not seen before.”

By focusing on female-initiated behaviors, the study highlights what is possible in evolution’s toolkit, sparking new questions about the complex dynamics of communication in the animal kingdom.

Yun Ding is an assistant professor in the Department of Biology in the School of Arts & Sciences at the University of Pennsylvania.

Dawn Chen is a postdoctoral researcher in the Ding Lab in Penn’s School of Arts & Sciences.

Minhao Li is a Ph.D. student in the Ding Lab in Penn’s School of Arts & Sciences.

Other authors are Ian P. Junker, Fabianna I. Szorenyi, Guan Hao Chen, and Arnold J. Berger of Penn Arts & Sciences and Aaron A. Comeault and Daniel R. Matute of the University of North Carolina at Chapel Hill.

This work was supported by the National Institutes of Health (grants R35GM148244 and R35GM142678).