Ex Vivo Brain Imaging in Drosophila

Assemble the Imaging Chamber

• 1. Assemble a Warner RC-26G recording chamber on top of a glass microscope slide.

• 2. Spread a thin bead of vacuum grease around the bottom edges of the chamber, using the corner of a Kimwipe, a syringe needle, or forceps, and seal the chamber to the slide.

• 3. Use hot glue to secure corners of the chamber to the slide and prevent lateral movement.

• 4. Using a scissors or razor blade, cut small, ∼5-mm pieces of PTFE tubing to use as connectors between the inlet/outlet tubing on the chamber and sections of tubing running through a dual-channel peristaltic pump.

  • These will function as quick disconnects.

• 5. Insert the small PTFE tubing pieces into the tubing in the inlet/outlet of the RC-26G.

• 6. Cut one section of Tygon tubing long enough to run from an artificial hemolymph reservoir (beaker or bottle) through the peristaltic pump to the imaging chamber when it is mounted on the microscope stage.

• 7. Cut another section of Tygon tubing one size larger in diameter than the inlet tubing and long enough to return from the outlet of the imaging chamber through the peristaltic pump to a large waste container in a suitable location.

  • This prevents saline overflow from the chamber.

• 8. Load these sections of tubing into the peristaltic pump rotor such that the saline flows toward the chamber in the inlet tube and away from the chamber in the outlet tube.

  • The saline level in the chamber is controlled by the vertical position of the outflow tube, and, if adjusted properly, the chamber will not run dry.

• 9. Make a Sylgard dish by filling a 50-mm Petri dish approximately halfway with Sylgard and allowing it to cure, following the manufacturer's instructions for Sylgard.

• 10. (Optional) Mount the glass slide with recording chamber into a 60-well overflow dish by using epoxy to fix four pairs of magnets on the underside of the slide and in complementary positions on the dish. Make a slide adapter by cutting a piece of polycarbonate to the size of a microscope slide (75-mm × 25-mm) using a table saw. Mount the polycarbonate slide adapter to the bottom of the overflow dish with hot glue or epoxy.

  • Alternatively, a microscope slide can be used, but it will not be shatter-resistant.

  • Note that if an overflow dish is not used, the slide will be mounted directly to the slide holder on the microscope stage.

Drosophila Brain Explant Preparation

• 11. Anesthetize a fly using a Flypad, Flowbuddy, and CO₂, or ice as follows:

  • Anesthetize a fly with a brief pulse of CO₂ (e.g., on a Flypad with a Flowbuddy).

  • Cool a fly on ice by placing the entire vial in an ice bucket until the flies are anesthetized (∼30 sec).

• 12. Fill the Sylgard dish about halfway with Drosophila artificial hemolymph.

• 13. (Optional) Handling the fly with forceps, dip it briefly in 70% ethanol to remove the wax on the cuticle and facilitate the dissection.

• 14. Transfer the fly to the Sylgard dish under the stereomicroscope and pin the thorax/abdomen down with minuten pins.

• 15. Using one pair of forceps, grab the fly head on one side (the cuticle near the eye is a good spot). Using the other pair of forceps, grab the other side, and begin to pull apart the cuticle.

• 16. Working bit by bit, remove pieces of cuticle (and the eyes) from each side of the head, exposing the brain.

  • The brain can be kept attached to the cervical connectives and body as long as is convenient.

• 17. Once the brain is exposed, remove the bright white tracheae, cleaning as much as possible from the exterior of the brain. Throughout the dissection, gently hold one side with one pair of forceps while working with the other pair. If the brain is still attached at the end of the dissection, remove it from the body by pulling on the cervical connectives with forceps.

• 18. In preparation for imaging, fill the imaging chamber with ∼1 mL of artificial hemolymph. Place one new 12-mm glass coverslip into the imaging chamber and press it to the bottom of the hemolymph.

• 19. Using a pair of forceps, carefully transfer the brain from the Sylgard dish to the imaging chamber (lift the brain, using the closed forceps like a spoon, rather than pinching it).

  • As it is transferred to the imaging chamber, it will tend to float initially due to surface tension.

• 20. Gently press the brain below the surface of the hemolymph with a pair of forceps. As it sinks, guide it down to the middle of the coverslip and orient it as desired (with the region of interest level and usually facing upward).

  • The brain will adhere to the surface of the coverslip without any additional adhesive coating.

• 21. Orient the brain and touch it down to the coverslip. Use a new coverslip for each experiment to ensure adhesion.

Imaging

• 22. Place the imaging chamber into the slide holder on the confocal microscope.

• 23. Connect the saline inlet and saline outlet tubes to the respective tubes on the peristaltic pump and turn on the pump (2 mL/min).

• 24. Monitor the saline flow closely for the first few minutes, ensuring that the saline outlet tube is pulling saline out of the chamber and that there are no overflows. Throughout the experiment, monitor the chamber for overflows.

  • See Troubleshooting.

• 25. Center the brain under the microscope objective and dip the objective into the saline (use a dipping objective, e.g., a 20×, 1.0-numerical aperture immersion objective).

• 26. (Optional) Use epifluorescence illumination to help locate the brain and zero in on regions of interest (ROIs) before switching to confocal or multiphoton imaging.

• 27. Change to image-acquisition mode and select the imaging settings using native image acquisition software for the microscope (e.g., Leica LAS X). Use an initial fast, low-resolution setting to help find the field of view, optimize the imaging parameters, and select ROI. For GCaMP, use a 488-nm laser (925-nm if using two-photon microscopy), with emission filters between 500 and 650 nm.

• 28. Adjust laser power to minimize photobleaching. Ideally use an 8- to 10-kHz resonant scanner to capture a time series of calcium responses to odors, but a standard scanner will work as well, with final image-acquisition frequency of 2–5 Hz.

  • Specific parameters depend on the signal intensity and on the physiological conditions of the neurons to study (ensure that there is enough temporal resolution to capture relevant changes in neural activity). For isolated brain preparations using pharmacological treatments, longer time series recordings and lower frame rates than those with in vivo approaches will be optimal (a recommended starting point is 8-min recordings at 1 Hz).

• 29. Once an ROI has been circumscribed and the desired plane is in focus, turn off the laser to avoid photobleaching. Prior to collecting the time series, select the final desired settings, including image resolution and line/frame averaging.

  • The specific parameters will depend on laser-exposure times; minimize light exposure while maintaining desired image quality. For example, with fast resonant scanning, line averaging (∼5×) often improves image quality appreciably without damaging the tissue as long as the laser power is kept minimal.

• 30. Collect a time series for long enough to collect a series of frames of prestimulus baseline fluorescence, deliver the stimulus and monitor the response rise and decay, and monitor the post-stimulus baseline.

  • See Troubleshooting.

Stimulus Application

• Stimuli can be provided to explant brain preparations in several ways. These include pharmacological approaches, with drugs applied either in the bath or locally via picoinjection or iontophoresis (Tomchik and Davis 2009). High-K⁺ saline can be bath-applied to depolarize neurons.

• 31. To apply high-K⁺ saline or drugs in the bath, move the inlet tube from the reservoir containing artificial hemolymph and place it into a 1.5-mL microcentrifuge tube containing the drug of interest dissolved in artificial hemolymph at the desired concentration (or the high-K⁺ saline).

  • The peristaltic pump can be stopped during the transfer or left on if the transfer can be done quickly. The latter has the advantage of introducing a small air bubble in the tube that allows the experimenter to know precisely when the drug reaches the imaging chamber.

• 32. Deliver the drug at the desired concentration for the desired time, and then transfer the saline inlet tube back to the saline reservoir (cleaning/wiping quickly with a Kimwipe if necessary).

  • A common procedure is to deliver 1 mL of drug/saline for 30 sec (flow rate of 2 mL/min).

• 33. If desired, measure the fluid dynamics of the chamber in test trials by running the stimulus procedure with an empty chamber and replacing the drug with fluorescein at a low concentration (followed by a long pulse for 10 min). Image the empty chamber using microscope settings equivalent to those for GFP/GCaMP imaging (488-nm excitation).

  • This allows the experimenter to determine the kinetics of drug application (and estimate the final/maximum drug concentration at the brain, relative to what was put in the inlet) in their system.

Data Analysis

• 34. As with analysis of in vivo imaging data (see Protocol: Imaging Olfactory Learning-Induced Plasticity in Vivo in the Drosophila Brain [Boto and Tomchik 2024]), draw an ROI around each neuron or neuropil for analysis, and export the average fluorescence intensity across this ROI over time.

  • This can usually be done in the native image-acquisition software for the microscope. Alternatively, export the raw images as time series (or an image stack) and analyze them with image-processing programs such as ImageJ.

• 35. Calculate the baseline fluorescence before stimulus application (F₀) and average the fluorescence across a 3 sec time window immediately preceding the stimulus (the F₀ window).

• 36. Normalize the change in fluorescence at each time point to this baseline using the following equation: (F_i – F₀)/F₀, producing ΔF/F. Depending on the temporal resolution of the data acquisition, use longer F₀ and response windows to accommodate the slower responses (e.g., that accompany bath application of pharmacological stimuli) (Tomchik and Davis 2009).

• 37. Calculate the response either as the maximum value in a response window, defined as an 8-sec window around the odor delivery, or as the area under the curve.

  • The choice of parameter to quantify depends on the experimental question—is the maximal response or total response over time of greater interest? The F₀ window and response window can be adjusted for different stimuli/paradigms if necessary.

• 38. Compensate for photobleaching, if present, by fitting a polynomial function to the time series trace (omitting the response window from the least-squares calculation to prevent the stimulus-evoked Ca²⁺ transient from affecting the fit) and subtracting it from the Ca²⁺ trace.