Ablation of a Neuronal Population Using a Two-photon Laser and Its Assessment Using Calcium Imaging and Behavioral Recording in Zebrafish Larvae

Summary

Here, we present a protocol to ablate a genetically labeled subpopulation of neurons by a two-photon laser from Zebrafish larvae.

Abstract

To identify the role of a subpopulation of neurons in behavior, it is essential to test the consequences of blocking its activity in living animals. Laser ablation of neurons is an effective method for this purpose when neurons are selectively labeled with fluorescent probes. In the present study, protocols for laser ablating a subpopulation of neurons using a two-photon microscope and testing of its functional and behavioral consequences are described. In this study, prey capture behavior in zebrafish larvae is used as a study model. The pretecto-hypothalamic circuit is known to underlie this visually-driven prey catching behavior. Zebrafish pretectum were laser-ablated, and neuronal activity in the inferior lobe of the hypothalamus (ILH; the target of the pretectal projection) was examined. Prey capture behavior after pretectal ablation was also tested.

Introduction

To understand how behavior arises from neuronal activity in the brain, it is necessary to identify the neural circuits that are involved in the generation of that behavior. At the larval stage, the zebrafish provides an ideal animal model for studying the brain function associated with the behavior because their small, transparent brains make it possible to investigate neuronal activity at a cellular resolution in a broad area of the brain while observing the behavior¹. Imaging of neuronal activity in specific neurons has become possible through the invention of genetically encoded calcium (Ca) indicators (GECIs) such as GCaMP². GCaMP transgenic zebrafish have proven to be useful for associating the functional neural circuit with behavior by conducting Ca imaging in behaving animals³.

While Ca imaging can demonstrate correlations between neuronal activity and behavior, to show causality, suppression of neuronal activity and testing its consequence(s) on behavior are important steps. There are various ways to achieve this: use of genetic mutation that alters specific neural circuits⁴, expression of neurotoxins in specific neurons^5,6, use of optogenetic tools such as halorhodopsin⁷, and laser ablation of targeted neurons^8,9. Laser ablation is particularly suited for eliminating activity in a relatively small number of specific neurons. Irreversible elimination of neuronal activity by killing neurons facilitates assessing behavioral consequences.

One interesting behavior that can be observed at the larval stage in zebrafish is prey capture (Figure 1A). This visually-guided, goal-directed behavior provides a favorable experimental system for the study of visual acuity¹⁰, visuomotor transformation^11,12,13, visual perception and recognition of objects^{14,15,16,17,18}, and decision making¹⁹. How prey is recognized by predators and how prey detection leads to prey catching behavior has been a central question in neuroethology²⁰. In this paper, we focus on the role of the pretecto-hypothalamic circuit formed by projections of a nucleus in the pretectum (nucleus pretectalis superficialis pars magnocellularis, hereafter, simply noted as the pretectum) to the ILH. Laser-ablation of the pretectum was shown to reduce prey capture activity and abolish neuronal activity in the ILH that is associated with the visual prey perception²¹. Here, protocols for performing laser ablation and testing its effect using Ca²⁺ imaging and behavioral recording in zebrafish larvae are described.

Protocol

1. Ablation of a Subpopulation of Neurons Using a Two-photon Laser Microscope 　

Note: If users plan on performing Ca imaging following ablation, use the UAShspzGCaMP6s line²¹. If users plan on performing behavioral recording following ablation, use the UAS:EGFP line, as the ablation of EGFP-positive cells is easier to perform than of GCaMP6s-expressing cells.

1. Begin by setting up the mating of a Gal4 line that labels specific neurons to be studied and UAS:EGFP or UAShspzGCaMP6s. Be sure to use the nacre background for both parents so that nacre homozygotes can be obtained.
   Note: Homozygotes of nacre contain no melanophores on the body surface (Figure 1B). In the present study, a Gal4 line gSAIzGFFM119B (which labels the pretectum) and another Gal4 line hspGFFDMC76A (which labels the ILH) are used²¹.
2. On the following day of the mating setup, collect the zebrafish eggs, and raise them to the larval stage at which point the neurons of interest will express fluorescent reporters.
3. Mount the zebrafish larva in 2% low melting point-agarose (Figure 1B).
   1. Begin by preparing 2% agarose stock solution: dissolve 2 g of low melting-point agarose powder in 100 mL system water (i.e., the water used to maintain adult zebrafish in a circulating water system)²² or E3 water²³, by bringing the water to the boiling point in a microwave, and mixing vigorously.
   2. Microwave the 2% low melting-point agarose stock until it boils. Pour 3.5-4.0 mL of the agarose onto the lid of a 6-cm Petri dish. Wait until the agarose cools so that it is warm to the touch before proceeding.
   3. Next, place a single larva (not anesthetized at this step) in the agarose. Keep adjusting the position and orientation of the larva so that it is upright, using a dissecting needle (Figure 1C), until the agarose solidifies to ensure the proper positioning of the larva.
      Note: The larva will continue to swim around while the agarose solidifies.
   4. Once the larva is positioned, allow 5 min for the agarose to solidify completely.
   5. Pour 5 mL of tricaine solution (0.4% at 1x working concentration) over the agarose.
      Note: To avoid movements by the larvae during laser ablation, the larvae must be kept anesthetized throughout the ablation procedure.
4. Ablate the larva using the bleaching function in a two-photon laser microscopy system.
   1. Place the agarose-embedded larva under a two-photon laser scanning microscope and start the microscopy system.
   2. Start the image acquisition software and click the "On" button on the "Laser" tab to turn on the laser (Figure 1D). Wait for the "Status" of the laser to change from "Busy" to "Mode-locked."
   3. On the "Channels" tab, set the laser's wavelength at 880 nm (maximum power: approximately 2,300 mW) for EGFP ablation, or at 800 nm (maximum power: approximately 2,600 mW) for GCaMP ablation.
   4. Select the 20X objective lens by manually moving the lens revolver. Click the "Locate" tab, click "GFP" to change the optical pathways, and directly view the fluorescence by eye.
   5. Locate the zebrafish larval brain at the center of view; then click the "Acquisition" tab to go back to the two-photon microscopy.
   6. As a record of the 'before ablation' condition, take a z-stack image with the 20X objective lens at fast scanning-speed (to avoid heat-damage). Later compare the images taken before and after ablation experiments (Figure 2A). To obtain a z-stack, click "Z-Stack" to select the z-stack option and expand the tab. Set the lower limit (click the "Set First" button") after focusing on the ventral end of the larval brain, and set the upper limit (click the "Set Last" button) after focusing on the dorsal most-surface of the brain. Then, click the "Start Experiment" button to run the z-stack image acquisition.
   7. Select the 63X objective lens by manually moving the lens revolver.
   8. Click the "Locate" tab to switch to the epi-fluorescent microscopy, locate the cells to be ablated at the center of view by eye, then click the "Acquisition" tab to go back to laser scanning microscopy.
5. On the "Acquisition Mode" tab, set the "Frame Size" at 256 x 256. Click "Live" and observe the neurons of interest.
6. Starting from the dorsal-most side, find a focal plane in which the cells to be ablated are in focus.
7. Mark a small, circular area about one third the cell's size in diameter on each neuron as a region of interest (ROI) using the "Regions" function.
8. Set the scan speed to 13.93 s/256 x 256 pixels (134.42 µm x 134.42 µm), which corresponds to a laser dwell time of approximately 200 µs/pixel, or 200 µs/0.5 µm. Also, set 4 repetitions ('Iteration cycle: 4'). Alternatively, optimize the number of iterations and scanning speed so that sufficient ablation is achieved.
9. Perform the 'Bleaching' function for ablation, in combination with the 'Time series (2 cycles)' to image the sample (before and after ablation) and 'Regions' (to set the ROIs) or equivalent functions in the software used (Figure 1D).
10. By comparing the first image (before ablation) and the second image (after ablation) in the Time series cycle, ensure that after bleaching, the fluorescence in the targeted cells decreases to that observed at the background level (Figure 2B).
11. If fluorescence is still present after ablation, increase the number of iterations.
12. Next, move the focal plane slightly deeper (i.e., towards the ventral side), and choose the next focal plane where un-ablated cells appear.
13. Perform the bleaching function as described above (step 1.9). Repeat these steps until all the cells in the neural structure of interest have been ablated.
14. After going through all the focal planes covering the neural structures to be ablated, check the cells for abolished fluorescence. If some cells are found to still be fluorescent due to insufficient ablation, laser-irradiate them again as described above.
15. If the larva needs to be recovered from agarose after laser irradiation, do not try to force it out of the agarose because this might damage the larva. Instead, carefully make tiny cuts in the agarose around the larva perpendicularly to the surface of its body. Then, wait for the larva to swim out of agarose on its own when the anesthetic wears off.
16. Proceed to the next larva for ablation or perform control experiments using another population of neurons that are uninvolved in the behavior under investigation.
    Note: Here, as a control, olfactory bulb neurons in UAS:EGFP; gSAIzGFFM119B larvae were laser-ablated using the same protocol described above (Figure 2A, right).
17. Allow several hours or 1 day for recovery before proceeding to Ca imaging (Section 2) or behavioral recording (Section 3). During this recovery time, check the larval health (i.e., normal spontaneous swimming, no apparent heat-damage around the ablated area, etc.) and remove larvae showing any signs of poor health.

2. Calcium Imaging to Record Prey-evoked Neuronal Activity in the Pretectum-ablated Zebrafish Larvae

1. To prepare a recording chamber, put a secure-seal hybridization chamber gasket on a glass slide, with the adhesive side facing on the glass, and remove the top seal.
   Note: This creates a small recording chamber that is 9 mm in diameter and 0.8 mm in depth (Figure 3A).
2. Next, place a nylon mesh (size of opening: 32 µm) on a 50-mL tube that has the bottom cut open. Use the cap to attach the mesh on the tube (Figure 3B).
3. Prepare the paramecia (which is the prey to be used as the visual stimulus for larva).
   Note: Prepare the Paramecium culture (steps 2.3.1 and 2.3.2) beforehand.
   1. Obtain paramecia seed culture from commercial sources or academic bio-resource centers²⁴ in advance.
   2. Culture the paramecium for several days in water containing autoclaved rice straw and a pellet of dry beer-yeast²² (Figure 3C).
   3. Pour the paramecium culture through 2 sieves successively (one with a 150-µm aperture, followed by another with a 75-µm aperture) to remove debris from the culture.
   4. Collect paramecia in the flow-through from the previous step on a nylon mesh (13-µm aperture).
   5. Re-suspend the paramecia on the nylon mesh into a glass beaker using a squeeze bottle of system water (Figure 3D).
4. Rinse the mesh (Figure 3B) in system water 2x (Figure 3E).
   Note: The purpose of this step is to remove any possible odorants and tastants contained in the paramecium culture medium, as the goal of the present study is to observe visually driven neuronal activity.
5. Place a zebrafish larva in the tricaine solution to anesthetize.
6. Mount a single larva in 2% low melting-point agarose.
   1. First, microwave 2% low melting-point agarose and add 1/10 volume of 10x concentrated tricaine to the agarose.
   2. Place one drop of the tricaine-containing agarose onto the recording chamber (Figure 3A).
   3. After confirming the agarose is not too hot, place the anesthetized larva (from step 2.5) into the agarose, immediately removing any excess agarose so that the surface of the agarose looks slightly convex.
   4. Quickly orient the larva into an upright position with a dissecting needle (Figure 1C).
      Note: These procedures must be completed within several seconds before the agarose solidifies.
   5. Once the larva is properly positioned, allow 5 min for the agarose to solidify completely. Then, pour a drop of 1x tricaine solution on the agarose.
7. Remove the agarose around the head of the zebrafish larva and make space for the paramecium to swim using a surgical knife. To do this, first, make a few small cuts in the agarose (Figure 3F). Then, carefully remove the excess agarose by gently pouring system water on the chamber.
   Note: At this step, the tricaine will be washed out.
8. Next, put one paramecium in the recording chamber to serve as a visual stimulus.
   Note: When the paramecium swims near the head of the zebrafish larva, it will evoke a neuronal response (Figure 3G).
9. Place the recording chamber under an epi-fluorescence microscope equipped with a scientific CMOS camera (Figure 3H) and select a 2.5X objective lens (Figure 3I).
10. Collect time-lapse recordings of the GCaMP fluorescence signals.
    1. Start the image acquisition application for the equipped camera.
    2. Click "Sequence Pane" and select "Hard Disk Record" on the pane. In the Scan Settings section, set the "Frame Count" to 900 or any other total frame number desired.
       Note: If available, use time-lapse recording software with a module (here, "Hard Disk Record") that enables efficient saving of the data onto a hard drive.
    3. In the "Capture" pane, set the exposure time at 30-ms (i.e., 33.3 fps), and Binning 2 x 2.
    4. On the microscope controlling touch panel, choose a filter set for GFP, as the GCaMP has similar excitation/emission spectra to GFP.
    5. Click "Live" on the Capture pane to locate the larva in the camera view and focus (Figure 4A), and then click "Stop Live" and click the "Start" button. Save the movie in .cxd or any other format that holds information about the acquisition condition.
11. After the recordings, analyze the GCaMP6s fluorescent signals (Ca signals) by obtaining mean pixel values for the ROI set on a brain structure in the time-lapse movie using the free software Fiji²⁵.
    Note: Fiji is a version of ImageJ with many useful plug-in's preinstalled and can read .cxd format image files with the Bio-Formats plug-in.
12. If the larva move slightly during the recording, perform image registration by running the TurboReg plug-in²⁶ in Fiji. From the menu, select 'Plugins', 'Registration', and 'TurboReg'.
13. Consider how to analyze the data according to the objective of the study. The following is an example.
    1. Calculate the F/F0 (fluorescence intensity at each time, divided by the fluorescence intensity at rest or before any Ca transient has occurred) as follows:
    2. Set ROIs on the pretectum or the ILH using the ROI Manager: in the menu, select "Analyze," "Tools," "ROI Manager", and "Add" to the list of the ROIs (Figure 4A).
    3. Calculate the mean pixel values for the ROIs (i.e., fluorescence intensities of GCaMP6s) by selecting on the ROI Manager panel: "More," "Multi-Measure", and "OK." Save the Results as a spreadsheet file.
    4. As the signals are noisy, average the signals for 11 time-points (moving average) using an application that can handle spreadsheet data.
    5. Divide F (fluorescence intensities over time) by F0 (fluorescence intensity at the basal level) to calculate the normalized fluorescence intensity. As an example of data visualization, changes in normalized GCaMP6s fluorescence intensity in the pretectum and the ILH are color-coded and mapped on the paramecium trajectories (Figure 4B - D).

3. Assessment of Behavioral Consequences Following Laser Ablation

1. Set up a behavioral recording system that consists of a stereoscope equipped with a video camera. Use a camera with an appropriate image acquisition software, commercial or custom-made (Figure 5A).
2. Optimize the lighting condition so that the paramecium can be detected by image processing. For example, use a ring white LED light with a spacer (Figure 5B).
   Note: The distance from, and the angle of, the LED affects the appearance of the sample (i.e., bright in dark background or dark in bright background) and are, therefore, critical for successful image processing. Determine the best height of the spacer (Figure 5C).
3. Place a zebrafish larva (recovered from the laser-ablation experiment described in Section 1) in a behavioral recording chamber (which was attached to a glass slide) with the original top seal removed (Figure 5D).
4. Collect 50 paramecia from the culture (step 2.3.5) using a micropipette and place them in the recording chamber.
5. Place a cover slip on top of the recording chamber. Put a small drop of water on the cover slip before placing on the water surface of the recording chamber to avoid introducing air in the chamber.
   Note: In this step, some water with paramecia will overflow from the recording chamber. The remaining number of paramecia in the recording chamber will be approximately 30-40.
6. Place the recording chamber (containing a larva and paramecia) in the lighting system (Figure 5D).
7. Start the time-lapse recording at 10 fps for 11 min (6,600 frames).
8. (Optional) For each larva that was tested for behavior, observe the fluorescence of the laser-ablated areas by fluorescent microscopy to confirm the effectiveness of the laser-ablation (i.e., absence, or significantly reduced number of, fluorescent cells in the laser-irradiated area).
   Note: Even after irradiation, some fluorescent cells can still be observed owing to later onset of Gal4 gene expression in cells that differentiated after ablation²⁷.
9. After recording the larva's behavior, to compensate for the lack of uniformity in the background, perform a pixel-by-pixel division of each frame by an average image in Fiji: click 'Process', 'Image Calculator', 'Operation: Divide', '32-bit(float)result', and 'OK'. To make an average image, click 'Image', 'Stacks', 'Z-Project', 'Average intensity', and 'OK' (Figure 5E, F).
10. Count the number of paramecia left in the chamber by clicking 'Analyze', 'Analyze Particles', setting an area range that covers all the paramecia, checking the 'Show:Masks' checkbox, and clicking 'OK'. The 'Show:Masks' option will provide an extracted paramecia image (Figure 5G), with a list of the number of particles (paramecia) in each frame.
11. Plot the number of paramecia left in the recording chamber over time. Because the estimates of the number of paramecia are noisy, we recommend performing a moving average on each point in a range of 600 frames (1 min) on the spreadsheet.

Results

Specific neurons were genetically labeled with either EGFP or GCaMP6s, whose expression were driven in Gal4 lines. A Gal4 line gSAIzGFFM119B was used to label a nucleus in the pretectal area (magnocellular superficial pretectal nucleus), and a subpopulation of olfactory bulb neurons. Another Gal4 line, hspGFFDMC76A, was used to label the ILH. We laser-ablated the pretectal neurons bilaterally (Figure 2A left panel) and also ablated neurons in...

Discussion

Although the two-photon laser has an excellent spatial resolution to specifically ablate individual neurons, great caution should be taken to avoid any undesired damage on the brain tissue owing to heat. The most important step in the ablation experiment is to determine the optimal amount of laser irradiation. Insufficient irradiation fails to kill the neurons. Too much irradiation will heat-damage the surrounding tissue, which will result in undesired effects. The optimal range of laser irradiation (areas of the ROIs, n...

Materials

Name	Company	Catalog Number	Comments
NuSieve GTG Agarose	Lonza	Cat.#50080	low-melting temperature agarose
6 cm petri dish	FALCON	Product#:351007
dissecting needle	AS ONE Corporation	Cat. No. 2-013-01	https://keystone-lab.com/en/item/detail/404142
LSM7MP	Carl Zeiss		two-photon laser scanning microscope
W Plan-Apochromat 63x/1.0	Carl Zeiss		63X objective lens
Imager.Z1	Carl Zeiss		an epi-fluorescence microscope
ZEN	Carl Zeiss		Image acquisition software for confocal microscopes
Secure-Seal Hybridization Chamber Gasket, 8 chambers, 9 mm diameter x 0.8 mm depth	Molecular Probes	Catalogue # S-24732	Used as a recording chamber in Ca imaging
Imageing Chambers	Grace Bio-Labs	CoverWell Imaging Chambers PCI-A-2.5	Used as a behavioral recording chamber
surgical knife	MANI	Ophthalmic knife MST15
ORCA-Flash4.0	Hamamatsu Photonics	model:C11440-22CU	a scientific CMOS camera
HCImage	Hamamatsu Photonics		image acuisition software
Hard Disk Recording module	Hamamatsu Photonics		An software module that enables saving the movie files onto a hard disc drive in a short time
SZX7	Olympus		stereoscope
DF PL 0.5X	Olympus		objective lens for SZX7
Point Grey Grasshopper3 4.1 MP Mono USB3 Visio	FLIR Systems, Inc.	Product No. GS3-U3-41C6NIR-C	CMOS camera
XIMEA xiQ camera	XIMEA	Product No. MQ042RG-CM	CMOS camera
a ring LED light	CCS	Model: LDR2-100SW2-LA	White LED
Nylon mesh 32µm	Tokyo Screen	N-No.380T	http://www.tokyo-screen.com/cms/sta20347/
Nylon mesh 13µm	Tokyo Screen	N-No. 508T-K	http://www.tokyo-screen.com/cms/sta20347/
Metal seive 150 micron aperture	Tokyo Screen		http://www.tokyo-screen.com/cms/sta20341/#ami
Metal seive 75 micron aperture	Tokyo Screen		http://www.tokyo-screen.com/cms/sta20341/#ami
EBIOS	Asahi Food & Healthcare, Co. Ltd.		dry beer yeast
LabVIEW	National Instruments		an integrated development environment for programming
Mai-Tai HP	Spectra Physics		two-photon laser