Using Light Sheet Fluorescence Microscopy to Image Zebrafish Eye Development

The overall goal of this experimental protocol is to image eye development of the Zebrafish embryo in the light sheet microscope and process the acquired multiview data set in Figi via the multiview reconstruction application plug in. This method can help you to understand morphogenetic events in animal development. The main advantage of this method is that you can observe morphogenetic events in the developing in tact embryo with high spacial and temporal resolution.

The major advantage of light sheet microscopy are the low phototoxicity of the method, the fast acquisition timing with which you can image the sample and that you can image the sample from various angles, so called views. Capturing the sample from different views allows to us overcome the degradation of the signal and the set axis. These different images can then be used and registered using fluorescent microspheres around the sample.

We can reconstruct the three dimensional volume reaching the isotropic resolution. Although we demonstrate the use of light sheet flourensence microscopy for Zebrafish eye development, this method can be also used to answer various questions in developmental biology and neuroscience using various model organisms. To begin, screw the detection objective into the microscope while keeping the illuminating objectives covered.

Then, uncap the objectives and carefully slide the sample chamber into the microscope. Secure the chamber by tightening the screw. Then, connect the temperature probe and peltier block to the microscope.

The two tubes that circulate the cooling liquid for the peltier block are compatible with both connectors and can work in either orientation. Next, use a syringe to load the chamber through the syringe with solution up to the upper edge of the chamber windows and check for leaks. Now, switch on the microscope, the computer, and the incubation.

Start the microscope operating software and set the incubation temperature to 28.5 degrees Celsius. While the temperature requilabrates for an hour, prepare the sample. Begin with preparing the agarose.

Melt a prepared one milliliter aliquot of 1%low melting point agarose in E three medium at 70 degrees Celsius. After about 15 minutes, transfer 600 microliters of molten agarose into a fresh 1.5 milliliter tube and add 250 microliters of E three medium, 50 microliters of 0.4%MS 222, and 25 microliters of vortex fluorescent bead stock solution. Place the tube in a heating block at about 39 degrees Celsius to let the agarose approach its gelling point.

Next, push Teflon tipped plungers down to the bottom of one millimeter inner diameter glass capillaries. Prepare five of these. Now, briefly vortex the agarose mixture and then transfer five embryos with the minimal amount of liquid into it.

Then, insert a capillary and aspirate one embryo head first. The head must enter before the tail and there cannot be any air bubbles. Draw up enough solution so there is about two centimeters of agarose above the embryo and one centimeter of agarose below it.

Draw one embryo into each capillary before proceeding. This step requires some practice to master. Prepare for it using non valuable embryos.

When the agarose solidifies completely, store the samples in E three medium by suppporting the capillaries to a beaker wall with plasticine with the bottom opening in solution. Now assemble the sample holder. First, insert two plastic sleeves of the right size against each other into the stem.

The slits on the sleeves must face outwards. Next, attach a clamp screw. And insert the capillary through the holder until the black color band becomes visible on the other side.

Avoid touching the plunger. Then tighten the clamp screw. Next, push a centimeter of agarose out from the capillary and trim it off.

Then insert the stem into the sample holder disk. Before loading the sample disk, check that the stage is in the upper most position in the control software. Then, use guiding rails to glide the holder and sample down into the microscope.

Now, image the chosen sample as described in the text protocol. The data processing is done in Figi. It is a distribution of ImageJ.

Before starting, Figi should be updated. Save the acquired image data set into a single folder. The intermediate files of the image processing will be saved to the same folder.

Now, start the multi view reconstruction application. Then, define a new data set. Select the option Zeiss Lightsheet Z.1 Dataset LOCI Bioformats and create a name for the output data file.

Then select the first CZI file of the data set. The meta data will be loaded automatically. In the image meta data, double check that the number of the angles, the channels, the illuminations, and the voxel size are all set correctly.

Upon pressing OK, a log window, the view set up explorer window, and a console window open up. The view set up explorer provides a user friendly interface. Select the files that need to be processed and press the right mouse button.

A new window will open showing the different processing steps as they are executed. In the top right corner are the info and save buttons. Info displays a summary of the content of the saved data file and pressing save will save the processing results.

The program does not save the results of the processing steps otherwise. To resave the entire data set, first, select all the files by pressing Ctrl+A and using a right click. Then select Resave Dataset and As HDF5.

A confirmation of the resave action must then be checked off. After a few minutes, for each time point, the program will indicate when the resave is complete. Then, proceed with the detection of interest points in the data.

Select all the time points by pressing Ctrl+A and right-click. To select, detect interest points. From the options, select difference-of-gaussian.

And select, down sample images prior to segmentation. In the next window, choose the detection settings. From the subpixel localization options, select 3-dimensional quadratic fit.

Then, in the interest point specification, select interactive. For the down sample XY option, select match Z resolution. And for down sample Z, select 1x.

Then, execute the operation by selecting the compute on CPU Java option. From the pop up window, select one view for testing the parameters. Once the view loads, adjust the brightness and contrast by pressing Ctrl+Shift+C on the keyboard.

Also, for bead detection, toggle the look for maxima green option. Now observe the segmentation as green rings around the detections. Adjust the difference-of-gaussien values for sigma one and threshold parameters to optimally segment the beads with the minimal number of false positives.

Each bead must only be detected once. After determining the optimal parameters, press done. After the computer loads, each individual view of the time point and segments the beads, the program outputs a log file with the number of beads it detected per view.

There should be between 600 and several thousand beads per view. Otherwise, the detection parameters need tweaking. This is a tricky process.

The detection pyramid is rarely accurate the first try and the resets can vary between the time points of one movie. Relaxing the fresh will detect all the real beads and allow more false positives can help. If no combination of the pyramid does works, then a new data set using different fluorescent beads has to be acquired.

Now proceed with registering the detections and making a multi view deconvolution from the stack. LSFM is an ideal method for imaging developmental processes at multiple scales from inter cellular structures to cells and entire tissues for long periods of time. Very fast inter cellular events can be captured at high resolution, such as the growth of microtubules in neuro progenitor cells.

From a single view movie, quantification of microtubule growth is possible. Intracellular structures can be followed over many hours such as the centrosome within retinal ganglion cells. Single cell behavior can be extracted from whole tissue images such as the trans location of a ganglion cell across the retina.

Multi view deconvolution creates one isotropic Z stack by combining several different views using the point spread function. This provides vastly improved images compared to single views or weighted average fusions. For example, the transformation of the optic vesicle into an optic cup can be studied from optical slices in different orientations through the data set shown.

Additionally, the overall good image quality, even deep inside the embryo is apparent. Such as when viewing optic cup morphogenesis. After watching this video you should have a good understanding of how to mount an image Zebrafish embryos in a light sheet microscope.

And how to process multi view data sets in Figi. Once mastered, it will take around two hours to start a time lapse experiment on light sheet microscope. The subsequent data analysis will take significantly more time.

It might be beneficial to use a computer cluster for that task. More practical tips, trouble shooting advice, and links to other resources can be found in the text protocol. There has to be enough storage capacity and a clear data processing pipe plan available ahead of the experiment.

If you have any problems, questions or future requests, please refer to the respective feature website or file issues on the respective github repositories.