In-vivo and Ex-vivo Calcium Imaging of an Olfactory Circuit Neuron in the Third-Instar Drosophila melanogaster Larva

This protocol uses GLUture tissue adhesive to enhance the ability in in-vivo and ex-vivo calcium imaging. This allows researchers to understand the calcium dynamics in an olfactory circuit neuron in Third-Instar Drosophila larvae. Simplicity and cost efficiency are the two main advantages for this technique, making experiments more reliable and accessible in neurophysiological research.

Prepare two feeding chambers by placing a small square of chem-wipe in six centimeter Petri dishes. For starved conditions, add 350 microliters of distilled water. For non-starved conditions, add 350 microliters of 0.2 molar sucrose solution.

Transfer an equal number of washed larvae into each feeding chamber. Allow the larvae to feed on sucrose, non-starved, or distilled water, starved, for two hours at room temperature On a 24 by 50 millimeters, 1.5 millimeters thick microscope cover glass, create a well, using petroleum jelly dispensed from a syringe. Place a small drop of GLUture topical tissue adhesive in the center of the cover glass.

Spread the GLUture evenly into a thin, uniform layer, using a glass rod. Carefully transfer a single larva onto the GLUture using a brush or thin wire. Gently press the ventral side of the larvae onto the adhesive.

Allow the GLUture to dry, ensuring the larvae is fully immobilized. Once the larvae is immobilized, immerse the larva in 100 microliters of imaging buffer. Place the cover glass onto a Leica DMi8 inverted microscope, connected to a Yokogawa CSU-W1 spinning disc confocal scanner module, and a CCD camera for calcium imaging.

Prepare a microscope cover-glass with GLUture for calcium imaging, as described in steps 3.2 to 3.3. Complete the dissection as described by Ishimoto and Sano, 2018. Using a P-10 micro pipette, carefully aspirate the dissected brain with five microliters of dissection solution from the Petri dish.

Slowly expel the brain onto the GLUture. Before the GLUture dries, orient the brain in a dorsal side up position with the two brain lobes facing down and the dorsal ventral cord facing up. Using a P-200 micro pipette, transfer 100 microliters of calcium imaging buffer over the immobilized larvae brain on the cover glass.

Open VisiView for calcium imaging, using Leica DMi8 inverted spinning microscope. Capture images using a 10X/1.4 any objective with filters, switching between the GFP laser and RFP laser. Exposure values range from 100 to 1000 milliseconds and the gain is set at 50%of the corresponding exposure value.

A bidding factor of two is applied during image capture. Identify the neuron of interest by searching for the tdTomato signal with the RFP laser. Once the region of interest is identified, switch to the GFP laser to capture GCaMP6f signals.

Ensure both tdTomato and GCaMP signals colloquialize before capturing stack images of both channels. Capture two separate time series recordings of tdTomato and GCaMP6f success signals simultaneously at a rate of 0.5 frames per second, for a total of two minutes. Import the tdTomato and GCaMP6f signal stacks into ImageJ.

To identify the correct region of interest, merge the tdTomato and GCaMP6f signals to confirm colocalization. After verification, split the tdTomato and GCaMP6f channels. Import the custom ROI-based macros into ImageJ.

Select the GCaMP6f stack and hit run for calcium analysis. The macro first generates maximum intensity projections, performs motion correction, using stack reg plugin, and applies bleach correction. Motion correction was carried out by defining a broad rectangular region of interest, such as the brain lobe on each frame.

The rectangular box must cover the entire ROI in all frames to ensure accurate motion correction, while retaining local, spatial, and ROI shape across all frames. Each ROI, the macros calculated one, the intensity across all frames, two, the baseline fluctuation difference, such as the difference between the maximum and minimum intensity values during the first 10 frames, and three, the maximum frame to frame intensity change, Delta F.ROIs showing intensity fluctuations less than 10 units were eliminated. The final dataset was organized into four columns, one, time and frames, two, sample conditions starved or fed, three replica ID, individual samples, and four, normalized intensity.

F norm value scaled from zero to one. Finally, data analysis and visualization were performed in R, using custom R scripts. The required packages, including ggplot2, dplyr, and readr were installed and updated prior to running the analysis.

The output included heat maps illustrating the normalized calcium intensity for the first 30 frames, 60 frames, and all frames at the time series recording, and box plots to show median of normalized intensities. The significance of normalized calcium intensity was assessed using the Mann-Whitney U test. Figure one is a schematic for in-vivo and ex-vivo calcium imaging setups.

Whole larval sample A, or larval brain sample B, was fixed on a thin layer of GLUture in solid blue, spread within a well of petroleum jelly, and round on a microscope cover glass. Samples were immersed in calcium imaging buffer, light blue, and placed for imaging on an inverted spinning disc confocal microscope. Time series of calcium fluorescence images of the samples were captured.

Schematic was prepared using BioRender. Figure two illustrates GCaMP6f fluorescence imaging. A, split images showing fluorescence in the 488 nanometer channel, GCaMP6f and green, and the 525 nanometer channel, tdTomato in red.

The merged image is shown in the right panel. B, example raw fluorescence traces over time for GCaMP6f in green and tdTomato in red in keystone LNs. Fluctuation in GCaMP6f fluorescence indicate calcium activity while the stable tdTomato signal is used as a control to identify keystone LNs.

C, representative GCaMP6f signal images for in-vivo top panels and ex-vivo bottom panel preparations. Non-starved samples are shown on the left, and starved samples are shown on the right. Figure three depicts GCaMP6f fluorescence measurements.

A, the top left panel shows a heat map showcasing the average temporal dynamics of calcium signals from keystone elements from non-starved, N=5, and starved larvae, N=5, in the in-vivo preparation whole larvae. The box plot in the top right panel compares normalized calcium signal intensity between non-starved in gray, and starved in white samples in the in-vivo preparation. Mann-Whitney U test, P is less than 0.001.

B, the bottom left panel shows a heat map showcasing the average to portal dynamics of calcium signals from Keystone LN from non-starved, N=5 and starved N=5 in the ex-vivo preparation dissected brain. The box plot in the bottom right panel compares normalized calcium signal intensity between non-starved, gray, and starved, white, samples in the ex-vivo preparation. Mann-Whitney U test B is less than 0.001.

To demonstrate our larval calcium imaging approach, we successfully applied it to both in-vivo and ex-vivo preparations of the drosophila larvae. In both preparations, we observed higher keystone LN activity and starved samples compared to non-starved samples. These higher keystone activity was clearly observed in heat maps, showcasing temporal representations of the responses over 60 seconds, and in box plots depicting average signal intensity values.

These results are consistent with the recent studies that demonstrate higher olfactory neuron activity in starved animals. As demonstrated in this video, GLUture tissue adhesive significantly reduces motion artifacts in both in-vivo and ex-vivo setups. This method could also be applied to study different kinds of stimuli, for example, external application of odors and internal superfusion of neuromodulators.