Suggestions on how to improve behavior at each step of the dissection/preparation process.

A healthy, intact, freely walking fly will walk at 10 mm/s or more, with rest periods clearly delineated from walking periods, so that the histogram of forward velocities is clearly bimodal (see e.g. DeAngelis et al. 2019, Figure 1). On a spherical treadmill setup, this should also be visible, as in these data from Gaudry et al. 2013, Supplementary Figure 1; note also that the distribution of lateral (sideways) velocities is fairly symmetric around zero:

Getting strong walking behavior in head-open flies can be much more difficult. A healthy, head-open fly might only reach 2-6mm/s, sometimes reaching 6-10mm/s.

Rearing#

Crosses - In general, you want to only be collecting the healthiest flies for your experiments.

  • To avoid over/under-crowding the vial, shoot for 7-10 females and 2-3 males per vial.
  • To ensure you are collecting the first flies to eclose (typically the healthiest), flip your cross to a fresh vial every 2-3 days.
  • To ensure the parents remain healthy, set a new cross roughly once a week.
    • You can set crosses more frequently of course, but this would mean more time fly-pushing for diminishing returns

Food - Flies reared on molasses food typically have more robust behavior than german food.

  • Anecdotally, some behaviors can be quite difficult to elicit unless the flies are reared on a specific food. It is worth trying a few different food options (molasses, german, corn, etc) when piloting new experiments to determine which is best for your behavior of interest.

Timing - Flies typically behave best at their subjective morning or evening (<4 h from light-dark switches), so it is best to set up crosses in both the 8-8 and 4-4 incubators.

Collecting#

CO2 vs Ice - The effects of anaesthetizing flies with CO2 can sometimes last up to 24 h. Flies collected on ice typically have better behavior than those collected on CO2.

Wings - Optionally, clipping the wings during collection can make it easier later when mounting the fly into the holder. It is also better than damaging the wings day of an experiment, as this can cause discomfort to the fly.

Housing

  • Low Density - Female flies for experiments are typically kept at low density (3-5/vial) prior to experiments.
  • Social Isolation - Male flies for experiments are typically kept single-housed (1/vial) prior to experiments.

Starvation - Some flies will only walk well when food deprived. To starve a fly, place it in a vial with a damp Kimwipe to avoid desiccation. Importantly:

  • The optimal period of food deprivation is genotype- and behavior-dependent
    • Note: in colder times, flies might be more unhealthy and shorter starvations periods (3-7 h instead of 16-24 h) tend to work better.
  • Also, food deprivation does not always improve behavior and can sometimes, in fact, worsen behavior!
  • Thus, some trial-and-error is required to determine the optimal conditions (if any) for your flies.

Fly Selection - Sometimes, a fly was never meant to be a good walker. To ensure that the flies that you are using have the best chance of being good walkers prior to mounting/dissecting, you should inspect their walking behavior in the vial. If you tap them down, which flies run to the top first? Which ones are running fastest? Which ones look bigger/healthier? Use those!

  • Thus, it is always helpful to collect more flies than you need for a given experiment.

Mounting#

Holders - Flies typically behave much better in the custom cone-shaped holders than the flat foils.

Body Position

  • The head should be snug against the front of the holder, but not squished. The angle of the head will depend on the cells of interest.
    • For flies that are mounted in holders for electrophysiology or imaging experiments, the angle of their head relative to their body can sometimes have a dramatic effect on walking behavior. All else being equal, the more the head is tilted forward (angled away from the body) and the more “neck” strain the fly experiences, the less likely you are to get good walking.
  • The thorax should be snug in and flush against the holder
    • Do not place the body too high, otherwise the fly cannot walk well. A good reference point is that the wings should be placed just above the holder, and the halteres should be just below.
    • Do not crush the body, otherwise the fly cannot walk well or will die. If you notice that the body is at all concave when placed in the holder, strongly consider trying a larger slot size.
  • The wings should be above the holder. If they are below, the wings can either physically get in the way of the fly’s walking behavior or the fly will be tempted to groom instead of walk.
  • The abdomen should be free to move and not in the holder at all.
    • You can even take a pair of blunt forceps to push the abdomen away from the holder to ensure that it is completely dislodged.

Glue

  • In general, most folks glue on both sides of the thorax and on both eyes
    • When gluing the thorax, the glue should naturally spread against the side of the body
    • When gluing the eyes, the glue should just barely fill the gap between the head and thorax. Avoid glue on the dorsal cuticle
  • Use as little glue as possible and glue one section at a time (e.g., left body, right body, left eye, right eye). Too much glue can either:
    • Result in glue dripping down onto the legs/face of the fly
    • Result in the fly overheating upon UV-curing, as the reaction is exothermic
  • Do NOT get glue on the abdomen (back of the slot), the fly typically will not walk if their abdomen is uncomfortable.
  • Important: After gluing, the fly should be flailing as vigorously as before
    • If you notice the fly now appears lethargic, or is twitching substantially (some is fine), or is cross-legged, I can almost guarantee you that that fly will not walk well once on the ball and you should toss it
    • This may indicate that you need to use less glue or take more time between sections
    • This may also indicate that your fly was not very healthy/robust to begin with, for instance, male and starved flies are much more sensitive to the curing reaction because they are smaller/more frail respectively

Proboscis

  • You can either sever the lateral muscles and glue the proboscis in place, or just glue the proboscis in place. Severing the lateral muscles can improve brain stability, however, it can also worsen behavior. So it is a trade off that should be considered. If the location of your cell body permits you to simply glue the proboscis without severing the later muscles - this is recommended.

Dissection#

With emphasis for getting good behavior for electrophysiology experiments:

  • The best use of your time is to sharpen forceps.
  • Remove just enough cuticle to comfortably expose the cell bodies of interest. Less is great.
  • When desheathing - be incredibly careful to not pull off any cell bodies. Air on the side of undershooting and take trips between the dissection scope and the rig to asses how well they are desheathing. 4-5 trips back and forth is not ridiculous. Contrary to what it may seem - the flies do not mind the travel between the scopes but will not run if too many cell bodies have been ripped off. Similar to the cuticle - only desheath what is absolutely necessary to patch the cells of interest. This is critical.
  • Muscle 16 is the bane of every Drosophila electrophysiologist’s existence. Use a hook to gently sever the muscle 16 but be extremely careful as to not rip off the antennal lobe neurons. These run laterally and are easy to knick (ask someone to draw them out for you if you are unsure). If you damage these neurons, the flies will consistently stop running in some 20 minutes.

On the ball#

Position - While the flies are surprisingly good at walking regardless of their positioning, it is important to position the fly well on the ball in order to maximize good walking behavior.

  • When looking at the fly from behind,
    • Adjust the y (left/right) position until the fly is centered above the ball
    • Adjust the roll and yaw axes until legs and body are evenly balanced with respect to the ball
  • When looking at the fly from the side
    • Adjust the x (back/front) position until the fly is centered above the ball
      • The fly should be able to rest their back legs on the ball, so you may need to push them forward just a bit if you notice that they are leaving their back legs in the air at rest
    • Adjust the z (height) position until the thorax is roughly 1 body-width (thorax) away from the surface of the ball
      • From behind, the legs should be just below 90-degrees when bent at rest
      • The fly should be able to dip its abdomen and touch the ball surface
      • The fly should not look like they are reaching for the ball at all
    • Adjust the pitch axis until the thorax is roughly parallel with the surface of the ball

Heat - Warm saline typically produces better walking. The saline can be warmed using a coffee-mug warming device under your saline flask, or by using a in-line heater with closed-loop temperature control (e.g., Warner TC-324C). Temperatures of close to or slightly greater than 30C tend to work well.

  • Heating the actual enclosure might be helpful as well. The objective->microscope acts as huge thermal sink that take quite a while to equilibrate to the saline, and getting to that equilibrium point might require setting the inline saline heater to an very high value (which will drive out all of the oxygen, if that’s a concern). LED displays tend to keep things toasty, however, so a dedicated heater might not be needed in that case.
  • Flies under the 2P scope are probably experience saline temperatures close to 30C. For reference on Berg-III, good walking behavior is observed when the saline temperature reaches 26-28C at the fly holder, which is sustained by the in-line heater with control temperature set to 40-41C. Note that the 2P room has room temperature less than 21C, which makes saline cool down fast.
  • There is anedcotal evidence that flies will walk “like gangbusters” if they spend a day or more at ~20.5C and are then shifted to a warm temperature for the experiment (per Stephen Holtz).

Warm up - When a fly is initially placed on the ball, it will typically make lots of yaw movements, with relatively few bouts of forward walking. If the fly is going to start walking well, this will typically emerge over a 10-20 min warm-up period.

  • For some behaviors (e.g., courtship), the fly may require 30-60min before he is ready to engage well.

Humidity - (the highest that is not experimentally problematic) helps. Using a terrarium humidifier that can be refilled outside of the enclosure is most convenient. 

Saline - For imaging/brain-sheathed experiments, bringing the saline to 75mM glucose can dramatically boost walking behavior. Unfortunately, these conditions do not work for ephys/brain-unsheathed experiments.

  • Anecdotally, there is evidence that glucose may push the fly into different internal states. For instance, flies are less likely to menotaxis in the presence of high glucose, so it may be that this addition is better for some but not all behaviors.

Other#

Genotype - A few genotypes are just very poor walkers. This is one reason to consider building some alternative genotypes for key experiments: one genotype may just walk much better.

  • Consider trying your experiment with flies that someone in the lab knows are strong walkers
  • Consider backcrossing your flies to a healthier line

Superstition - Maybe the flies are running because the molasses food is particularly good this week, or your dissections are really excellent. But maybe it’s because you decided to give your pet an extra cuddle in the morning or started eating a new cereal for breakfast….

Some notes from Kaun lab on behavior experiments:

https://www.kaunlab.com/behavior-blog