Finding the ‘sweet spot’: what hungry flies can teach us about human metabolism

From the Rajan Lab, Basic Sciences Division

Think back to the last time you agonized over the decision to make one last run at the all-you-can-eat buffet—or that time you forgot to pack a lunch—and you will have appreciated the value of nutrient sensing in your daily life. Dysregulated nutrient sensing (and subsequent feeding behavior) is at the core of numerous metabolic maladies in humans, including obesity. And while there is a considerable body of work investigating the brain circuits which underly abnormal feeding behavior, the means by which your body informs those brain circuits of your nutritional status are much less understood. Enter Drs. Kevin Kelly and Mroj Alassaf, two post-doctoral fellows in the lab of Dr. Akhila Rajan, who recently co-authored a study published in eLife reporting a new piece of the puzzle towards understanding diet-influenced feeding behavior. The Rajan Lab studies fat-to-brain communication in the fruit fly, Drosophila melanogaster, a classic model organism which continues to surprise with its relevance to human biology. By studying the fly fat body (its ‘fat sensing organ’) the group hopes to understand the molecular crosstalk between adipose and neural tissues and apply that knowledge to inform treatment of human metabolic disorders.

Their recent study was born out of a desire to explore the effects of long-term diet on fly feeding behavior; for 5 to 28 days, they provided flies with either a ‘normal’ diet or a high-sugar diet containing 30% more sucrose. After various stints on these diets, the researchers then starved the flies for 16 hours before giving them access to food and monitoring how often they fed following starvation. For a while, there were no surprises. Predictably, flies under both feeding regimens fed more following starvation than those fed ad libitum—a behavior which the authors call ‘hunger-driven feeding’. However, starting on day 14, something remarkable happened: flies under a high-sugar diet began to stop seeking food after starvation. Importantly, this loss of hunger-driven feeding at day 14 got more pronounced the longer they kept the flies on a high-sugar diet but was absent in flies fed a normal diet. Crucially, the authors were able to show that loss of feeding behavior was not simply a result of these flies having more stored energy. 

Armed with these findings, the team was faced with a tempting follow-up question: what was happening on day 14 in flies fed a high-sugar diet? To address this, they focused on insulin signaling, a major regulator of systemic metabolism in both flies and humans. Using several molecular readouts of insulin signaling in both the brain and the fat body, the authors confirmed that insulin signaling was defective in flies fed a chronic high-sugar diet. Particularly interesting among these readouts was the morphology of lipid droplets in the fly fat body, which appeared progressively more enlarged and misshapen the longer the flies were kept on the diet. Overall, these molecular markers revealed that sugar-overconsuming flies—like their human counterparts—developed insulin resistance which was likely linked to their altered feeding behavior.

An artistic interpretation of a fruit fly on a high-sugar diet.
An artistic interpretation of a fruit fly on a high-sugar diet. Image generated by the author using DALL-E 2 software.

To dig even deeper into the abnormal fat body phenotype, the team measured levels of major lipids in flies fed a standard and high-sugar diet. To their surprise, they found dysregulated levels of two types of phospholipids—these are the lipid molecules which make cell membranes and serve signaling roles in the body. Several earlier studies in humans had noted a link between insulin resistance and dysregulated phospholipid metabolism; what these studies lacked, however, was the ability to test for a causal link between the two. Armed with their behavioral paradigm and the power of fly genetics, the team embarked to test whether phospholipids control feeding behavior by genetically manipulating an enzyme called Pect, which catalyzes the rate-limiting step of phosphatidylethanolamine (a type of phospholipid) in flies. They didn’t stop there, though—with the appreciation of the fat body as a bona fide nutrient sensing organ, the team decided to manipulate Pect expression only in the fat body.

After checking for insulin resistance in flies with fat body-specific reduction in Pect, the team was in for a surprise—Pect knockdown recapitulated many of the same signs of insulin resistance seen in the flies fed a high-sugar diet. In the ultimate test of their model, the authors created flies which overexpress Pect in the fat body and put Pect over- and under-expressing flies through the same behavioral paradigm which inspired the study. Astonishingly, reducing fat body Pect expression abolishes the hunger driven feeding response—not only in flies fed a high-sugar diet, but also in flies fed a normal diet. Correspondingly, increasing fat body Pect expression in flies fed a high-sugar diet preserved their hunger driven feeding responses. Altogether, these results establish a causal link between fat body phospholipid metabolism and feeding behavior in flies.

If you find it hard to believe that altering the expression of a single enzyme in a single organ of a fly is enough to dictate its feeding behavior in response to starvation, you’re not alone. As Mroj noted, “We were pretty shocked to find that such a small perturbation could recapitulate features which are considered hallmarks of diet-induced obesity, but it really highlights the value of studying metabolism on the level of the whole organism.” One key detail which sets their study apart from others in the field is the length of time they chose to feed their flies. “When you consider that a fly lives on average around fifty days, a 28 day-long high sugar diet really represents a chronic dietary condition, which is surprisingly understudied in our field,” Dr. Rajan notes, “Without implementing this longitudinal approach, we would never have stumbled upon this behavioral effect.” Although such a long-term study presented some technical challenges, the team is hopeful that their approach better models obesity in humans and can be extended to test other potential disease-related candidates arising out of human metabolic studies.


The spotlighted research was funded by the National Institutes of Health and the Directorate for Biological Sciences.

Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium member Akhila Rajan contributed to this study.

Kevin P Kelly, Mroj Alassaf, Camille E Sullivan, Ava E Brent, Zachary H Goldberg, Michelle E Poling, Julien Dubrulle, Akhila Rajan. (2022) Fat body phospholipid state dictates hunger-driven feeding behavior. eLife 11:e80282