Protocol for intelligent high-content screening of zebrafish embryos on a standard widefield screening microscope
Ravindra Peravali1,*, Jochen Gehrig1,*, Stefan Giselbrecht2, Dominic Lütjohann1, Yavor Hadzhiev3, Ferenc Müller3, Urban Liebel1
1 Institute for Toxicology and Genetics, Karlsruhe Institute of Technology, Campus North, D-76344 Eggenstein-Leopoldshafen, Germany, 2 Institute for Biological Interfaces 1, Karlsruhe Institute of Technology, Campus North, D-76344 Eggenstein-Leopoldshafen, Germany, and 3 Department of Medical and Molecular Genetics, University of Birmingham, Birmingham B15 2TT, UK

*R.P. and J.G. contributed equally to this work.
DOI: 10.2144/000113782


The development of automated microscopy platforms has enabled large-scale observation of biological processes, thereby complementing genome scale biochemical techniques. However, commercially available systems are restricted either by fixed field-of-views, leading to potential omission of features of interest, or by low-resolution data of whole objects lacking cellular detail. This limits the efficiency of high-content screening assays, especially when large complex objects are used as in whole-organism screening. Here we demonstrate a toolset for automated intelligent high-content screening of whole zebrafish embryos at cellular resolution on a standard wide-field screening microscope. Using custom-developed algorithms, predefined regions of interest, such as the brain, are automatically detected. The regions of interest are subsequently imaged automatically at high magnification, enabling rapid capture of cellular resolution data. We utilize this approach for acquiring 3-D datasets of embryonic brains of transgenic zebrafish. Moreover, we report the development of a mold design for accurate orientation of zebrafish embryos for dorsal imaging, thereby facilitating standardized imaging of internal organs and cellular structures. The toolset is flexible and can be readily applied for the imaging of different specimens in various applications.






Fabrication of array of 96 agarose molds

* HINT: If standard 96 well plates are used proceed with step 11

* HINT: The silicone template is available upon request from the authors

1. Prepare a 10:1 mixture of silicone elastomer base component and curing agent of Sylgard 184 according to the supplier's protocol (Dow Corning).

2. Pour approximately 20 – 40 mL of this mixture into the polymethylmethacrylate dish with the 96 keel shaped cavities.

3. Evacuate the lid with uncured silicone within an exsiccator to get rid of entrapped air bubbles.

4. Cure the silicone >48h at room temperature on an even and horizontal plate or table. (Curing time can be reduced by heat cure.)

5. Slowly and carefully demold the cured silicone tool (can be improved by using appropriate and non-toxic/biocompatible releasing agents).

✋REST: The silicone tool is very robust and can be stored long-term at room temperature.

6. Boil 1% agarose in water until it is completely dissolved.

7. Pour a thin layer of agarose into a lid of a 96 well plate (any manufacturer). Make sure the lid is positioned horizontally.

8. Add the silicone tool to the agarose. Position it such that the grooves match the positions of the wells of the 96-well plate.

9. After solidification remove the silicone tool and remove the excess agarose at the circumference.

10. Remove the foil from a clear 96 well plate and press it into the agarose.

* HINT: A lid with the positions of the wells highlighted will greatly facilitate the positioning of the silicone tool (e.g., Corning, Amsterdam, The Netherlands, Cat.-No. #3598).

Preparation of embryos for high content screening

11. Begin with embryos at approximately 22 hours post fertilization (hpf).

12. Supplement the medium with 0.003% N-Phenylthiourea.

13. At 48 hpf, dechorionate the embryos enzymatically by removing most excess medium and adding of 0.5 mL of 10mg/mL protease from Streptomyces griseus Type XIV to the embryos.

14. Shake occasionally until almost all chorions are removed.

15. Wash 3x with 400 mL of water in a beaker.

16. Transfer embryos into a clean Petri dish.

17. Anesthesize embryos using 0.03% tricaine.

18. Optional: array the embryos in 96 well plates (e.g., U-bottom plates) and proceed with step 22.

* HINT: This step is an example for a procedure for long-pec stage embryos but can be easily modified.

Arraying of embryos

19. Fill 2ml of water (supplemented with 0.003% Tricaine) into a corner of the above prepared tool.

20. Pipette the embryos in a volume of 100l into the wells using a cut 200 µL tip.

21. Orient the embryos ventrally within the grooves under a stereomicroscope.

* HINT: Use a bent needle or similar to orient the embryos.

ATTENTION: Once the embryos are oriented, do not move the 96 well plate sitting in the agarose. Always move the tool by touching it at the lid.

Acquisition of pre-screen data

See also (Liebel, et al., (2003)(Gehrig, et al., (2009)

22. Load the arrayed embryos into the Scan^R screening microscope.

23. For each embryo, acquire 1 z-slice in the required channels (e.g., bright field and GFP).

Generation of a template XML-file

24. Change all objective settings to match the desired magnification.

25. Adapt exposure times.

26. Activate the interpolate focus function and collect a few reference z-positions.

27. Adapt the number of z-slices and slice distances.

28. Choose an appropriate destination directory.

29. Select 1 well and image it using the new settings in order to generate a new experiment_descriptor.xml.

ATTENTION: Make sure to use the very same plate definition that was used to generate the pre-screen data. Also, do not move the plate within the stage!

ATTENTION: Do not image more than one well for the template file generation!

Generating a data folder

30. Store the Pre-screen data and the template XML-file in the same data folder.

ATTENTION: The template XML-file must be named experiment_descriptor.xml!

Starting the graphical user interface (GUI) zfGUI

31. Start Matlab and change the directory to the location where the GUI package is installed.

32. Type zfGUI at the Matlab prompt to launch the GUI.

* HINT: If the GUI appears cluttered, maximize the GUI screen.

Selecting the embryo orientation (See Figure 1A (1), in the main text)

33. Choose either “lateral” or “dorsal” orientation depending on the experiment.

Figure 1. Implementation of algorithms as a graphical user interface. (Click to enlarge)

ATTENTION: For automatic detection, one of the two orientations must be chosen. For manual click detection, it is not required to choose any orientation.

Choosing offset parameters (See Figure 1A (2), in the main text)

34. Enter x-and y-offsets (in m) to stage coordinates, if required, to compensate for stage calibration issues.

* HINT: If nothing is entered, 0 is taken as the default.

Automatic detection (See Figure 1A (3), in the main text)

ATTENTION: If “Auto Detect Head (BF Only)” is to be used, put only the bright-field data in the data folder.

35. For bright-field head image detection, click “Auto Detect Head (BF Only).” For combined bright-field and GFP detection, click “Auto Detect (BF + GFP).

36. Choose the location of the data folder from the GUI.

37. After the analysis is done, 2-folders are created in the data folder. One is the “Output” folder and it contains the images of the analyzed data and a text file (“not_imaged.txt”) with filenames of embryos not expressing GFP. The other folder is the “xmlOut” folder which contains the XML files to be used for subsequent high resolution imaging on the ScanR

HINT: No text file will be generated if there is no GFP data to be analyzed.

Manual click detect (See Figure 1A (4),in the main text)

ATTENTION: Put either bright-field or fluorescence data along with the template xml file in the data folder!

38. If no automatic detection is required, click “Manual Click Detect.”

39. Choose the location of the data folder from the GUI.

40. The first image in the chosen folder is displayed. A question whether this embryo should be imaged further is prompted. If the embryo is not required for imaging, type ‘n’ and press ‘enter’. The next image in the folder is displayed and the process repeats.

41. If the embryo is required for imaging further, press ‘enter’. Cross-hairs will appear on the image. Click on the region of interest. A question if this is really the chosen region of interest will appear. If yes, press ‘enter’. This position is then registered. If no, type ‘n’ and press ‘enter’. The cross-hairs will appear again. This process repeats until the required position is correctly chosen. After this, steps 40 and 41 are repeated until all the images in the folder are processed.

42. The “Output” folder contains the resulting images. The “xmlOut” folder contains the XML files for subsequent high resolution imaging.

Execute Batch Job on the Scan circ;R

43. In the Scan^R acquisition software, select the “Operate” Option and choose “Start Batch.”

44. Load the XML files from the xmlOut directory and click on Start to start the high resolution screen.

* HINT: Some software versions of the Scan^R software do not allow for large numbers (for e.g., 96) of files to be loaded at once. If this occurs, choose half of the number of files first, the second half next, and then click on Start.

Reagents and Consumables

N–Phenylthiourea (Sigma-Aldrich, Taufkirchen, Germany)

Protease from Streptomyces griseus Type XIV (Sigma-Aldrich, Taufkirchen, Germany)

Tricaine (Sigma-Aldrich, Taufkirchen, Germany)

96-well U-bottom plates (e.g., Nunc, Roskilde Denmark, Cat.-No. 143761)

96 Well Microplate, PS,Clear®, Chimney Well (Greiner, Frickenhausen, Germany, Cat.-No. 655096)

Polydimethylsiloxane (Sylgard 184, Dow Corning, Wiesbaden, Germany)


Scan^R high content screening microscope (Olympus Biosystems, Hamburg, Germany).


zfGUI tool (downloadable from http://www.itg.kit.edu/liebel-lab-resources.php)

Windows operating system

Microsoft .NET framework 4.0 (http://www.microsoft.com/net/download.aspx)

Matlab with Image Processing toolbox

More Information
Do you have question about this protocol? Would you like more information about this protocol?
Submit your questions here and Biotechniques will forward your question and contact information from our database to an expert at
Yes, please forward my question, along with my name, title, company, and e-mail address to
BioTechniques will forward your question to the company, who will contact you directly by email.
Request more information
Name Question
Job Title