Video tracking fruit fly larvae

While the behavior of adult fruit flies (Drosophila melanogaster) is very interesting, we can also learn a lot from their larvae. Larvae at just a couple of days old (e.g. <6) are active enough for tracking and are especially popular in research on smell and eating behavior. On this page, you will find a couple of examples of how video tracking is used in studies to assess larva behavior. 

Smell is a complex sense

Fruit flies rely heavily on smell. It is not only important for locating food, but also used heavily in mating behavior. In the fruit fly brain there are many odorant receptors (ORs) expressed on a range of olfactory sensory neurons (OSNs), and it is thought that a specific combination of a subset of ORs is necessary to identify a specific smell. 

What is smelling (chemotaxis)?

Fishilevich and colleagues from the Rockefeller University in New York were interested in finding out the extent to which a single OR contributes to chemotaxis: smelling a certain smell. They identified 25 ORs in 21 OSNs in Drosophila larvae and were able to engineer larvae that ranged from missing one single OSN, or having just one OSN. 

Some smelling neurons are more important than others

Ablating a single OSN does not seem to affect odor detection overly much. However, some OSNs seem more important than others, and allow the fly to smell different odors. Other OSNs may not be sufficient alone for detecting odors, but do enhance odor detection if it is a secondary neuron, activated with others. The olfactory sense of fruit flies is thus complex, with different neurons having different abilities in different combinations.  

How can you tell if it is the smell?

So how can you actually tell if an animal, in this case a really tiny animal, is actually smelling something, and if it can tell the difference between different smells? The best way is to monitor its behavior. In this case, scents were introduced to a larva in a petri dish; 12 petri dishes were monitored simultaneously while EthoVision tracked the animal’s X-Y coordinates. 

Reaction to smell in high-resolution

The odor presentation method used by Fishilevich and colleagues (inserting the odor in a small cap at one end of the petri dish) is a common assay. It is possible that introduction of the odor at one end of the petri dish results in an odor gradient across the petri dish, but this also may not happen. 

To assess the gradient/concentration of odors presented, the same laboratory carried out another experiment. In this, the researchers assessed what concentration a larva would detect a certain odor, and if it was sensitive to both exponential and linear increases of this odor along an odor gradient line. 

Smelling across a gradient line

For this Matthieu Louis and his colleagues created an assay using 96-well plate lids. This way, they were not limited by deep wells, but the small rings of the lids made it possible to deposit odors in specific places across the arena. Three lids were stacked upon each other. The bottom lid remained empty, the middle lid was filled with agarose as a medium for the larva, and the top lid contained the droplets of odor, either a single drop or a row of droplets that increased in concentration from left to right either in linear or exponential fashion. 

Tracking the path larvae take

The researchers used EthoVision to track a single larva in each assay, and found that wild type Drosophila are indeed capable of detecting an increase in concentration. When an odor line was used with the highest concentration in the first drop, that then decreased, larvae returned to the highest concentration. Mutants that could not smell did not show any of this behavior. 

This publication is in the form of a JOVE visual article. In this video, Matthieu Louis and Silvia Piccinotti explain how their assay works. Please notice that this experiment used an older version of our current EthoVision XT software. With newer versions it is quite easy to track multiple subjects simultaneously. 

Two noses?

In further work, this research group investigated the navigation strategies and function of the ‘noses’ (two symmetrical dorsal organs) of the larvae using similar assays with well plate lids.  

Larvae navigation strategies

Wild type larvae showed signs of having a direct navigation mechanism, with a high turning bias: the direction of their turns in the paths towards the odor source were indeed correlated with the location of the odor. 


The olfactory sensory neurons are located in two dorsal organs of the larva. Louis discovered a left-right asymmetry in the functioning of these organs. When one of the two organs did not work in mutant larvae, they were still able to locate higher concentrations of the odor, but their paths meandered more. The mutant flies no longer displayed the direct navigation mechanism. Furthermore, how well they performed depended on which organs, left or right, was disabled. 

How fruit flies are relevant to obesity research

Obesity affects 10% of the adult population and can become a serious, even deadly problem. Some may wish there was a magic pill to inhibit overeating, and though many supplements are available from the local drug store, a true pharmacological treatment to regulate appetite has yet to be discovered. 

High-throughput drug screening

Gasque and colleagues from The Rockefeller University in New York have used Drosophila melanogaster larvae to screen a large range of molecules for potential food-intake modulation. They identified metitepine as a potent candidate, and confirmed its effects on all 5 serotonin receptors in a cell-based assay. 

Target identification

Next, they used 5 fly mutants (one line lacking each of the 5 serotonin receptors), and tested the specific effects of metitepine. Results from this experiment showed that the 5-HT2a receptor subtype was the sole receptor responsible for the feeding inhibition effect of the molecule. 

Feeding and locomotion

To assess feeding behavior, researchers looked closely at mouth-hook contractions, and measured the optical density of the gut to estimate food intake. Through these behavioral and internal observations, researchers discovered that metitepine was indeed making the larvae eat less. To make sure this was a direct effect of the drug, and not due to increased locomotion for example (which might have decreased food intake as a secondary effect), EthoVision XT was used assess the locomotion of the larvae.