How to Perform a Live/Dead Assay Quantification Using Fiji

 

Live/Dead assay is a very common cell staining procedure. Live cells are stained with calcein and generate green fluorescence upon the excitation of their cytoplasm. Dead cells are labeled with dye EthD which binds to their DNA and fluoresces red. You can read more about different types of viability assays here.

This guide will walk you through a general approach in fluorescent microscopy image quantification from Live/Dead assay to asses cell viability. This workflow is suitable for images of 2D and thin 3D controls captured using a fluorescent or confocal microscopy for samples stained with calcein/ethidium homodimer (EthD). While this approach can be used for 3D cell culture, the results are more reliable if the sample is relatively thin and there are not many cells overlapping in different focal points.

For this analysis, we will be using free image processing software Fiji, an extended version of ImageJ . It is available for Mac, Windows, and Linux platforms.

To follow along, you can download an example Live/Dead image here.

Fig 1. Microscope image of Live/Dead assay of neural stem cells. Green (calcein) stains for live cells, red (EthD) for dead cells.

Fig 1. Microscope image of Live/Dead assay of neural stem cells. Green (calcein) stains for live cells, red (EthD) for dead cells.

You can follow the steps below for the manual quantification, but you can also download our Fiji macro and install it by going to Plugins > Macros > Install… and then use it by selecting it from Plugins > Macros > “live dead quantification.” You might need to modify specific settings of the macro to suit your needs by going to Plugins > Macros > Edit and selecting our macro file. There are instructions for modifications included directly in the macro file.

File opening

The example file is an RGB .tif image exported from the microscope software. For single files like this, simply go to File > Open and select your file. 

If you have a .lsm file (a common format for ZEISS confocal microscopes), you can open it using the same method. However, the open file will be a composite image with three different channels in a Z-stack that you can cycle through. 

Fig. 2: RGB composite file. You can cycle through channels and z-stacks using sliders below the image preview.

Fig. 2: RGB composite file. You can cycle through channels and z-stacks using sliders below the image preview.

If you took a Z-stack image of your sample and exported the file as a series of separate images, you can import it to Fiji by selecting File > Import > Image Sequence… and opening the folder with the Z-stacks. Next, you will need to choose Sequence Options by selecting the number of stacks to import. If you want to import everything, just leave the values as default. 

Fig. 3: Image Sequence Options

Fig. 3: Image Sequence Options

Image processing

To process the image, we need to have it in the right format: single channel, 8-bit images. 

If you are working with Z-stacks from .lsm file, go to Image > Stacks > Z-project. In the Z-Project options window, choose all your stacks and Max Intensity as a Projection type. It will merge all the slices of the image to a single one with pixels containing the maximum value from all images in the stack at the given location. 

From this point, file processing will be identical for .lsm and .tif files. 

To quantify signal in green and red channels, we have to split them and treat separately. To do this, go to Image > Color > Split Channels. You will see three separate windows, each containing a separate channel of your image in a gray-scale. Each window will have its channel name indicated in the brackets.

Fig. 4: Channel splitting

Fig. 4: Channel splitting

Since we will be using only red and green channels, you can close the window with a blue channel.  

Now, make sure that your images are in 8-bit space. You can see it indicated above the image preview. To change it, go to Image > Type and select 8-bit. 

Cleaning up the image

A high quality image has proper illumination, focusing, a low background, and good intensity of the signal. However, the world is often imperfect, so you can clean up the image in this step. 

To eliminate some background noise, you can adjust the brightness and contrast by going to Image > Adjust > Brightness/Contrast. A new window will open with a histogram of pixel intensities and values to adjust. To remove some background noise, click and drag the Minimum bar to increase it until the image looks cleaner without losing any important signal.

To increase the intensity of your staining, you can drag the Maximum slider down. When you are satisfied with the results, click Apply. Repeat this procedure for both channels separately. Keep in mind that since the channels were acquired differently, they may require different values.

Be consistent with your adjustments when processing multiple images so as not to introduce any bias into the analysis. 

Image segmentation

Now it’s time for image segmentation that will be used to count live and dead cells. 

First, we need to convert the grey-scale image into a binary black and white image. To do this, select the green channel and go to Image > Adjust > Threshold and drag the top slider to adjust your selection. Selecting a correct threshold is very important. Too high will result in losing a lot of signal while too low will fuse cells close to each other, making counting more difficult.

Fig. 5: Adjust Threshold values using the sliders. Make sure you have the ‘Dark background’ checkbox selected. The red overlay shows the preview of what are you selecting.

Fig. 5: Adjust Threshold values using the sliders. Make sure you have the ‘Dark background’ checkbox selected. The red overlay shows the preview of what are you selecting.

Here you can see examples of what’s a good and a bad threshold selection.

Fig. 6: Examples of different thresholds. Too high (left), too low (center), correct (right).

Fig. 6: Examples of different thresholds. Too high (left), too low (center), correct (right).

After adjusting the threshold, hit ‘Apply’. As a result, you will have a black and white image, where the signal should be white and background black. Repeat the procedure for the red channel as well. 

Next, we need to split the separate cells that appear as one. You can do it by selecting your image and going to Process > Binary > Watershed. It will create thin separation lines between what software detected as separate objects.

Image prepared in this way can be subjected to automatic counting. From the Analyze menu select Analyze Particles…

Fig. 7:

Fig. 7:

Here, you can select the inclusion criteria for your analysis. If there is still a lot of noise in the image after processing, you can filter it out by limiting the size of the particles selected. If you don’t know the minimum size of the cells, leave the size value as 0-Infinity and follow the next steps to find that out. 

Select the ‘Add to Manager’ checkbox and hit ‘OK’. 

The software will create Region of Interests (ROI) around each white spot on the image. By going to Window > ROI Manager, you will be able to see all the separate ROIs and select them. To determine the smallest size of the cell, find a white spot on the image that is the smallest but is still a cell, select its number in the ROI Manager and hit Measure in the ROI Manager window.

Fig. 8:

Fig. 8:

In the Results window, you can see Area column with a value for the measured ROI. You can use this as your bottom threshold for size selection in the previous step for Analyze Particles. 

If you can’t see Area in Results window, go to Analyze -> Set Measurements… and select Area checkbox. 

To redo Analyze Particles, select all ROIs in the ROI Manager and delete them by hitting the Delete button. Then go to Analyze -> Analyze Particles and change size according to your area measurement (you can choose a value a bit smaller than the result of ROI measurement). 

To find out how many live cells there are, follow these steps for the green channel. In the ROI manager, you will be able to see how many particles were detected. You can simply note the number or select all ROIs by clicking on one of them and pressing Ctrl+A (Cmd+A on Mac) and clicking on Measure button in the ROI Manager. After that, Result windows will display a table with the number of ROIs and different measurements that you can select in Analyze -> Set Measurements… such as Area and Shape descriptions. 

You can save the content of the result window as .txt file or simply copy and paste it to the Excel sheet. 

Repeat particle analysis for the red channel. Note that dead cells will be smaller than live ones, so your area limit should be modified accordingly. 

To calculate the percentage of live cells, divide the number of spots detected in the green channel by the total number of green and red cells in the image and multiply by 100%. 

To analyze the next image, remember to delete all ROIs from the previous measurement. 


If you need any further assistance with Live/Dead assay or image quantification you can contact us at [email protected]. Happy bioprinting!