Tools

Semi-automated analysis of podosome reformation using ImageJ

Analysis of podosome dynamics can yield important insights on the impact of novel podosome components or regulators. Primary human macrophages constitutively form numerous (>100) podosomes per cell, and thus present as an excellent system for the study of podosome dynamics.
Here, we describe assays that enable the measurement of podosome dynamics, namely reformation of podosomes, in fixed cells. The assay is based on the key influence of Src kinase on podosome formation and turnover, and use the general Src kinase inhibitor PP2 as a tool to disrupt podosomes (Linder et al. J. Cell Sci.,2000), which enables the analysis of synchronised reformation of podosomes.
Subsequent software-based image analysis facilitates quantification of cells and podosomes and allows statistical analysis, yielding parameters such as the number of podosomes per cell and podosome density.

Gallery of confocal micrographs from macrophages
stained for F-actin using Alexa488-labeled phalloidin to label podosomes (green), stained with CellMask to highlight individual cells (red), after application of the described ImageJ macro for cell (C), and after using the ROI mask to correlate podosomes (black dots) with individual cells (red outlines and numbers; D). Images show untreated cells. Note that use of the ImageJ “ROI manager” tool allows the combination or deletion of ROIs in case of oversegmentation. (i.e., ROI 7 in D (centre) is obtained by merging 2 segmented objects belonging to a single cell. White scale bar = 10 µm. (Cervero, P et al., Adhesion Protein Protocols, 3rd edition in the Methods in Molecular Biology Series, 2013) Click image to enlarge.

The analysis is based on the processing of images acquired in two channels (in this case red = cell body; green = podosomes), followed by merging the respective regions of interest (ROIs), yielding three different values: A) area covered by each single cell, B) number of podosomes in each single cell, C) density of podosomes (here defined as number of podosomes/100 µm2 of cell covered area).
In the first step, images acquired in the “cell channel” (HCS CellMask, Red stain) are processed in order to get a final mask corresponding to all whole cells present in the respective field of view. For this purpose, it is important to reduce the background noise as much as possible, to ease the subsequent segmentation step, which enables recognition of individual cells as objects by the software. All settings can be adapted depending on the user needs and the quality of the staining.
The following steps (in Italics) have been recorded in a macro.txt using ImageJ (http://rsbweb.nih.gov/ij/), therefore can be copy/pasted in a .txt file and then opened with ImageJ or carried out one by one manually (using the respective tools).
Clearly all the steps can be fine tuned, depending on the microscope settings and/or the user needs, and the basic imaging principles can be used with any imaging software.
It’s very important to calibrate the scale depending on the microscope settings, before applying any imaging tool !

Cell mask macro steps:
run(“8-bit”);
run(“Smooth”);
run(“Gamma…”, “value=0.40″);
run(“Median…”, “radius=4″);
run(“Gaussian Blur…”, “sigma=4″);
run(“Make Binary”);
run(“Invert”);
run(“Watershed”);
run(“Analyze Particles…”, “size=150-Infinity circularity=0.30-1.00 show=Masks clear exclude add”);

Once the identified objects (cells) are recorded in the “ROI Manager”, it is possible to modify them.
This includes combining two segmented objects belonging to one single cell in case of oversegmentation (select the ROIs that need to be merged, right-click and select the “OR (Combine)” function followed by “Add”; this will create a single new ROI that will include the previously selected ones, which have to be deleted individually).
Moreover you can now set the values for calculation, for example cell area (Analyze>Set Measurements…>check only “Area”), then manually select all ROIs and click on “Measure”.
The obtained area values can be saved in a spreadsheet.
Once the ROIs corresponding to the cell positions are recorded, open the “podosome channel” (phalloidin stain) of the same image and run a series of actions which will allow highlighting and better definition of podosomes as individual, countable objects.

Podosome macro steps:
run(“8-bit”);
run(“Smooth”);
run(“Subtract Background…”, “rolling=5″);
run(“Gamma…”, “value=1.30″);
run(“Convolve…”, “text1=[-1 -1 -1 -1 -1\n-1 -1 -1 -1 -1\n-1 -1 30 -1 -1\n-1 -1 -1 -1 -1\n-1 -1 -1 -1 -1\n] normalize”);
run(“Make Binary”);
run(“Watershed”);
run(“Analyze Particles…”, “size=0.10-50 circularity=0.80-1.00 show=Masks display”);
roiManager(“Show None”);
roiManager(“Show All”);
n = roiManager(“count”);
for (i=0; iThe final instructions highlighted in bold at the end of the “Podosomes” macro are intended to count automatically all the objects (podosomes this time) in each single “cell” ROI (previously detected and recorded in the ROI manager). The “Result” window will then show the number of podosomes detected in each cell ROIs at the end of the list (keeping the same numbering order saved in the previous “Cell mask” macro).
After applying for a single picture these two macros, it will be possible to save the number of podosomes detected in each identified cell. These values, in combination with the values for cell area, can be used to calculate podosome density (here defined as number of podosomes/100 µm2 of cell covered area).

More detailed informations and instructions about this method are provided in the following paper :
“Cervero, P et al., Adhesion Protein Protocols, 3rd edition in the Methods in Molecular Biology Series, 2013” (ed. A Coutts), Springer.