Procedure
- Place small water bath in the TC hood. Fill with water and set to 38.5ºC. Put media and agarose (if previously made) in water to warm.
- Prepare sterile stock agarose solution (3.2%) using Sea Plaque Low Melt Agarose (Lonza Cat # 50101)
- Combine 3.2 g agarose with 100 ml of ddH2O in small glass bottle
- Autoclave and place in water bath in TC hood to cool to 38.5ºC
- Note: can store stock agarose at 4ºC for short amount of time. Microwave to re-melt. Do not re-use extensively.
- Prepare base agarose layer (0.8%)
- For 1 6-well plate combine 1.75 ml of stock agar solution with 5.25 ml of growth media
- Pipette to mix and aliquot 1 ml per well. Immediately rock plate to completely distribute the agarose
- Cool plates for ~5-10 min at 4ºC to solidify agarose. Do not stack plates when cooling or they will not cool evenly
- Return plates to TC hood and warm to room temp
- Prepare cells
- Trypsinize and count cells.
- Resuspend cells at 1.28e4 cells/ml in 5 ml of growth media. This will be enough for three replicate wells with each well containing 1e4 cells/well. 64,000 cells total per 5ml.
- Prepare upper agarose layer with cells (0.9%)
- Note: Make sure agarose has cooled to 38.5ºC to avoid burning the cells.
- Add 1.4ml of stock agar solution to each 5 ml sample of cells
- Pipette gently to mix and immediately overlay 1 ml of cell/agar mixture into three separate wells containing the solidified base layer.
- Cool plates for ~5-10 min at 4ºC to solidify agarose. Do not stack plates.
- Return plates to TC hood o Add 1 ml of growth media to each well
- Incubate plates at 37ºC and 5% CO2 for 10 to 20 days, monitoring for colony formation. Incubation time will depend on the growth rate of the cells.
- Replace media every 4-7 days. Be very gentle when removing and replacing media so as to not disturb the agarose layer.
- Staining colonies
- Remove media o Add 1 ml of PBS containing 4% formaldehyde and 0.005% crystal violet to each well (Mix 19.9ml PBS + 100 µl 100X crystal violet stain)
- Incubate 1 hr minimum
- Pipette off staining media and either reuse or dispose in hazardous waste container
- Imaging colonies
- Take 4 photographs of each well (one per quadrant) using the dissecting microscope
- Note: Once you have set the focus and light settings, do not change them. Adjust the microscope and lighting to avoid shadowing on the edges of the images. The image should have the same relative background throughout.
Quantifying colonies using ImageJ
- Create and print out a thumbnail contact sheet of all your images. This is most easily accomplished on a Mac by opening all the images using Preview, then switching to thumbnail view. Perform a screen capture and copy to a powerpoint. Printout the powerpoint. Use this contact sheet to determine which images to discard based on artifacts in the image.
- Copy all the images that you will use for counting to a new directory. IMPORTANT: Do not use the original images when analyzing with ImageJ because you can accidentally permanently change the images. Always work with a copy of the images.
- Determine the optimal settings for "threshold", "size", and "circularity" by testing various values using your positive and negative control images as well as a few experimental images.
Threshold values
- You will set both a minimum and a maximum threshold and the range is 0 to 255.
- open an image in imageJ and convert to 8-bit (Image -> type -> 8-bit)
- Invoke the threshold function (Image -> Adjust -> Threshold). ImageJ will open a window with thresholding slide bars. All pixels that are above your maximum threshold will be false colored red. Experiment with different threshold values. The best setting should show colonies as red pixels and now background red pixels. An example could be min = 0, max = 150. Test out your best settings on several images to make sure there are no red pixels highlighting objects that do not appear to be colonies and that what appear to be colonies are red pixels.
- Once you have determined the optimal threshold values, enter them in the ImageJ macro to be used for processing all the files (See TKS Batch Count Colonies Macro below).
Particle size and circularity
- You will set a pixel-squared value for the boundaries of what should be considered a colony. This will be a minimum pixel-squared size and a maximum pixel-squared size (e.g. 50-5000). Only particles with this number of pixels-squared will be counted. Generally you will set a range that is around 50 pixels to infinity, unless you want to exclude very large colonies. Try multiple values for several images to see what colonies are excluded/included. Pick a value that excludes small dots that you co not consider to be colonies.
- You will also set a minimum and maximum "Circularity" parameter. The circularity of the particle in the image is calculated by ImageJ using a geometric formula, where 1.0 indicates a perfect circle and 0.00 is the opposite (whatever that is). Generally start with a range of 0.25-1.00. If you have lots of strange looking things that you want to exclude, increase the minimum value to 0.5 or higher.
- To play around with various particle sizes and circularity parameters do the following:
- Open an image in imageJ and convert to 8-bit (Image -> type -> 8-bit)
- Invoke the threshold function (Image -> Adjust -> Threshold...) and set the threshold value based on what you determined above.
- Convert the image to binary (Process -> Binary -> Make Binary)
- Invoke the analyze particles function (Analyze -> Analyze particles...). A window will open. Fill in the Size (pixel^2) window and the circularity window with your guess at the best values. Select "Outlines" from the Show drop down window. Check the "display results", "Clear results", "Summarize", and "Exclude on edges" boxes. Click OK. ImageJ will essentially erase your image, replace all particles that it counted with an outline and a sequential number inside each outline. A "results" window will open up that gives you measurements of each particle that it counted.
- Repeat the above step on several images until you are satisfied that you have the optimal particle size range and circularity range. Then enter these values in the ImageJ macro to be used for batch processing (See TKS Batch Count Colonies Macro below).
Counting colonies using an ImageJ macro
Run the macro "TKS_Batch_Count_Colonies". I have written a short macro that will open all .tiff files in a folder, analyze the particles, and report a summary of particle count, size, and area for all the .tiff images in the folder. The macro will change the images in the folder, so you can see what colonies were counted. Click here to see a text file of this macro (TKS_Batch_Count_Colonies.txt). Copy and paste this to a text file and put the file in your own ImageJ macro directory, then change the macro by changing the threshold, particle size, and circularity parameters in the macro with the ones that you determined above. Place a copy of all .tiff files to be analyzed in a separate folder, invoke the macro, select the folder, and run the macro. The summary window will appear after you run the macro and a log window indicating the number of images analyzed.
Copy the contents of the summary file to an Excel spreadsheet to analyze the data. To determine if differences between colony count, total area, or average particle size are significant use the Student's T-Test with two-tails and unequal variance.