Olfactory Behaviors Assayed by Computer Tracking Of Drosophila in a Four-quadrant Olfactometer

The overall goal of this setup is to assay olfactory behaviors of flies in a four quadrant olfactometer using a provided automated fly tracking software script. This method can help address key questions regarding how olfactory neural systems detect the external world, and how the brain decodes and directs behavioral decisions. The main advantage of this technique is that it's versatile and robust, and allows repeat testing of olfactory responses by population of flies in a large experimental arena.

For this protocol, prepare five odorant chambers that consist of a plastic outer container, glass inner container, and a custom-made PTFE lid insert. The insert is an original container lid with the central part removed, and two attached one way valves. Use four of the odorant chambers for solvent controls, and one chamber for a test odorant.

Make sure each chamber is clean, and then load an odorant container with one milliliter of solvent or odorant dilution. Then, check the position of the container in the plastic chamber, and ensure that no liquid has been spilt. Secure the lid using the screwtop and an O-ring on the PTFE insert to prevent leaks.

Rear the flies on standard cornmeal medium. In a standard bottle, place 30 males and 30 females, and allow eggs to be laid for five days at room temperature. From such bottles, collect newly eclosed flies.

For each experimental cohort, collect 25 male and 25 female flies under brief carbon dioxide anesthesia and let the cohort age on standard fly medium for two to four days. 40 to 42 hours before the experiment, transfer the cohort to a vial with about 10 milliliters of one percent agarose gel for humidity without food. Starvation helps to increase the whole motion, and more than 90%of the flies should survive the starvation process.

If more than 10%die, which can happen in a weak genotype, try using a one day starvation for all the flies in the experiment. At least five minutes in advance, switch on the temperature controller and set it to 25 degrees Celsius. Then, connect the odorant chambers to the arena via a plastic tube.

Check the airflow rate in each quadrant of the arena using an electronic flow meter. Make sure that the control and odorant air streams are both running at 100 milliliters per minute. Flies are very sensitive to airflow, and it is critical that the airflow for each quadrant is verified before each experiment.

Now, clean the arena and the glass plates using 70%ethanol. Wipe all the parts down two to three times, and allow them to fully air dry before proceeding. Next, clamp the glass plates to the arena.

Then, transfer the flies into the arena without anesthesia. Using gravity, let them enter through the hole in one of the glass plates, and then cover the hole with a circular mesh. Now, place the fly loaded arena into the light-tight chamber.

Then, connect the four control air streams to each arena corner. Close the door of the chamber and let the flies acclimatize to the new environment for 10 to 15 minutes. After the acclimation period, run a five to 10 minute control experiment in which flies are exposed to four control air streams.

It is essential to now analyze the data immediately. If the flies are not evenly distributed in the arena, the arena must be reset. It is critical to test the behavior of control flies to just clean air.

This will quickly verify that all background conditions are normal. Leaking light, temperature imbalance, tilted arena, or an odor contamination can all cause problems. If necessary, discard the flies and clean the arena again.

If an odorant contamination is suspected, replace all the tubing. Continue repeating the control test with new cohorts until the animals show no preference for any of the four quadrants. Then, connect the test odorant chamber to the setup using a three way valve, or reconnect the tubing and run a test experiment for five to 10 minutes.

For longer experimental recordings, rapidly stop and restart the tracking program every 20 minutes, or the data files will be too large. Between tests, discard the flies, clean the arena and glass plates with 70%ethanol, and replace the connector tubes. Keep the dry airflow going to continually flush the system.

For efficiency, it is good to have two arenas so cleaning can be done during runs. If several experiments are wrong on the same day, take extreme care to ensure that no odor is left in the system from a previous run. This is normally not a problem with low concentrations of odorant or with CO2, but for highly concentrated stimuli, up to a 24 hour gap between experiment runs may be needed.

For the data analysis, use the provided custom-made Matlab routine to convert the data into a Matlab format. Load the raw data file made in real time by the tracking software, and create a spacial mask that follows the contours of the arena. Apply the mask to the raw data to remove all the data points that fall outside of the arena, as they represent noise.

Then, remove all the data points that move at less than 0.163 centimeters per second for longer than three seconds. This data is likely to be noise, or non-moving flies. Now, visualize the remaining data points by plotting all of them through the recording periods.

The data can also be viewed as single trajectories. Odor boundaries can be quickly estimated based on the fly behavioral data. Then, calculate the attraction index, and make statistical tests as described in the text protocol.

25 male and 25 female flies were tested in the behavioral apparatus. In response to dry air, flies explored the entire arena with no preference for any quadrant. The line shows the tracked trajectory of one fly.

When apple cider vinegar was introduced into the top-left quadrant air stream, most of the flies were attracted to that odor quadrant. When a repellant odorant, 10%ethyl propionate, was introduced in the top-left quadrant, that quadrant was avoided. The attractive index of that quadrant scored 0.68.

An attractive index is the most widely-used index in olfactory response measurement. A value above or below zero indicates attraction or repulsion. The script compensated for slight variations in arena position between trials.

Data points that might have been noise or non-moving flies were removed. While the test can be run for any duration, it is interesting to view changes in the cohort's movements across 10 second bins to appreciate odor habituation and other phenomena. After watching this video, you should have a good understanding of how to effectively run and analyze olfactory behaviors in a four-field assay.

Once mastered, each olfactory experiment can be completed in about an hour, and we routinely run four to six experiments during one day.