top of page


Unveiling the Puppet Master: Parasitic Behavioral Hijacking in the Zombie Fruit Fly System

ANDRÉS MUEDANO SOSA, Harvard College '27

THURJ Volume 14 | Issue 2


It unfolds like a scene from a horror movie. A hungry fruit fly (Drosophila melanogaster) wanders in search of food, unaware of the fungal spore that has landed on its back. Over the next few days, the fungus (Entomophthora muscae) will infiltrate the fly’s body, breach its blood–brain barrier, and penetrate its brain tissue, eliciting dramatic behavioral changes. Finally, at dusk on its last day of life, the now zombified fly will climb to a high point along a vertical surface (a behavior known as summiting), secure itself to the surface via its proboscis, raise its wings so as to expose its dorsal abdomen, and die. In the following hours, the fungus will create canon-like structures along the fly’s exposed back—structures from which new fungal spores will be ejected. These spores will land on other unsuspecting flies, thereby initiating another cycle of infection.

Unraveling the mysteries of the E. muscae–D. melanogaster zombie fruit fly system remains an ongoing effort, with many lingering questions. How did the parasitic relationship between the fungus and fly first evolve? What molecules drive the fly to start climbing, specifically at sunset? And, perhaps most importantly, how is a parasite able to control the behavior of its host? The newly opened Elya Lab in the Department of Molecular and Cellular Biology, led by Assistant Professor Carolyn Elya, Ph.D., is committed to finding the answers to these questions.

From the Backyard to the Bench

Elya first encountered zombie flies as a graduate student in the laboratory of Michael Eisen, Ph.D., at the University of California, Berkeley. At the time, she was conducting research on the relationship between a fly’s microbiome and behavior. To obtain data from wild flies, she would frequently set traps in her backyard and screen for infected samples. One day, she found odd-looking flies with fungal spores accumulating along their backs. Intrigued, she collected them, sequenced their genetic material, and subsequently confirmed them to be infected with E. muscae.

Scientists have long known about the ability of E. muscae to infect and manipulate flies, with the first description of the fungus dating back to 1855 (Elya et al., 2021). However, the specific biological processes underlying fly infection and behavioral manipulation remain elusive. Seeking to address this, Elya changed trajectories for her doctoral research and focused on the development of a laboratory system to study the parasitic relationship between E. muscae and D. melanogaster. At Berkeley, she characterized the fly’s transcriptional response to the fungus and provided evidence that the fungus infects the fly’s brain, thus paving the way for the research she would subsequently conduct as a postdoctoral fellow and, now, assistant professor at Harvard. With a multidisciplinary approach integrating genetics, biochemistry, neuroscience, and microbiology, Elya and her team are now working to understand E. muscae’s pathogenic and behavior-manipulating abilities, guided by the following questions.

E. Muscae invades the brain. But does it have to?

Imaging experiments have demonstrated that E. muscae infiltrates D. melanogaster’s central nervous system 48 hours after exposure—perhaps an unsurprising finding given the observed behavioral effects of the infection (Elya et al., 2018). To infiltrate the brain, E. muscae disrupts the permeability of D. melanogaster’s blood–brain barrier—the membrane controlling which substances can pass from the fly’s “bloodstream” into the brain (Elya et al., 2018). In this sense, brain invasion appears to be a hallmark sign of fruit fly infection by E. muscae. But is this brain invasion necessary to trigger summiting behavior? Initially, it might seem self-evident that, without access to the brain, the fungus would not be able to affect fly neurons and thereby manipulate behavior. However, the recent discovery that transfusing hemolymph (the invertebrate equivalent of blood) from summiting flies to uninfected flies induces summiting-like behaviors suggests the possibility of a mechanism other than brain invasion (Elya et al., 2023). It is plausible that summiting is prompted by a metabolite present in the fly’s hemolymph, either secreted by the fungus or synthesized as part of the fly’s immune response. Such a metabolite could traverse the blood–brain barrier, triggering summiting without the fungus needing to invade the brain directly. To test this hypothesis, research assistant Charlie Heacock is currently investigating whether transgenic fruit flies bearing a mutation that fortifies the blood–brain barrier are less prone to summit. As such, Heacock’s research holds the potential to elucidate the mechanisms underlying summiting initiation, shedding light on the fascinating interplay between parasitic manipulation and host physiology.

Zombie flies will summit. What prompts them to?

The observation that summiting can be induced solely through hemolymph transfusions suggests that certain substances in the hemolymph trigger this behavior (Elya et al., 2023). While the identity of these factors remains unknown, Elya and her team have pinpointed the specific neurons they affect. Through imaging experiments, they found that fungal cells specifically accumulate in the pars intercerebralis (PI), a brain region that projects to the corpus allata (CA), an endocrine gland in the brain. Following this observation, they theorized that summiting might be driven by the release of juvenile hormone (JH), a locomotion-inducing molecule synthesized in the CA. Elya and collaborators tested this hypothesis by inhibiting the production of JH, observing that this treatment greatly reduced summiting behavior in infected flies (Elya et al., 2023). Next, they performed an experiment through which they disrupted candidate neurons using a potent neurotoxin and found that silencing PI neurons resulted in summiting inhibition. Together, these findings led to the theory that CA-mediated release of JH, driven by the activation of PI neurons, is what drives summiting behavior.

Based on these results, summiting factors likely induce changes in behavior by activating PI neurons, ultimately leading to increased locomotion. To determine the identity of these factors, Harvard GSAS student Julius Tabin is currently conducting assays to find what molecules in summiting fly hemolymph appear to activate PI neurons. Identifying these factors will lead to a better understanding of the neural mechanisms behind summiting behavior. Furthermore, studying how the sequences coding for these factors have changed over evolutionary time might shed light on how the relationship between E. muscae and D. melanogaster first emerged.

E. muscae is itself infected by a virus. What is the virus doing?

Elya previously conducted RNA-sequencing experiments to characterize the transcriptome of E. muscae cells and pinpoint the genetic mechanisms driving the infection process. However, the findings of these experiments proved to be startling: many of the identified RNA sequences could not be mapped to the genome of the fungus, meaning that the fungus could not have possibly synthesized them. Seeking to trace their origin, Elya compared the sequences to those readily available in online databases, soon finding a perfect and surprising match—a virus (Coyle et al, 2018). In a subsequent experiment, Elya’s collaborators confirmed the presence of viral RNA in E. muscae cells, both in culture and during fruit fly infection, providing evidence that the fungus is itself infected by a virus (Coyle et al, 2018). These findings only lead to further questions. Is the virus the puppet master of the infection, using the fungus as a mere carrier? Or, alternatively, is the fungus somehow weaponizing the virus, making use of it to successfully infect the flies? By studying how this virus contributes to E. muscae’s pathogenic abilities, Elya and her team aim to unravel the mechanisms and dynamics of a possibly tripartite biological interaction.

Zombiologists at Large

Studying the relationship between E. muscae and D. melanogaster offers a captivating glimpse into the intricate interactions between pathogens and their hosts. From the initial infiltration into the fly’s body to the induction of summiting behavior, each step of this macabre yet fascinating biological system presents a new layer of complexity begging to be unraveled. Through innovative research and interdisciplinary approaches, the Elya Lab is working hard to elucidate this complex symbiotic relationship, characterizing the relationship between host physiology and infection, identifying key molecules underlying observed behavioral alterations, and studying the involvement of additional biological agents in the infectious process. Through these concerted efforts, the Elya Lab’s research holds the potential to illuminate fundamental biological processes driving parasitic behavioral hijacking.


Coyle, Maxwell C. et al. (2018). Entomophthovirus: An insect-derived iflavirus that infects a behavior manipulating fungal pathogen of dipterans. bioRxiv, 371526. https://doi:10.1101/371526.

Elya, Carolyn et al. (2023). Neural mechanisms of parasite-induced summiting behavior in 'zombie' Drosophila. eLife, 12, e85410.

Elya, Carolyn, Henrik H. De Fine Licht (2018). Robust manipulation of the behavior of Drosophila melanogaster by a fungal pathogen in the laboratory. eLife, 7, e34414.

Elya, Carolyn, Henrik H. De Fine Licht (2021). The genus Entomophthora: bringing the insect destroyers into the twenty-first century. IMA fungus, 12(1), 34.
bottom of page