Automated Analysis of Intracellular Phenotypes of Salmonella Using ImageJ

This fluorescence microscopy method quantifies salmonella phenotypes inside intestinal epithelial cells in an automated way, allowing the study of salmonella on large numbers of strains with scalable resolution. Our method has the advantage of being quantitative, high-throughput, and automated at the same time, relying on both an area and a single-cell analysis. The intracellular behavior is central to the host-pathogen interaction of many bacteria.

Our method can be easily adapted to bacterial pathogens other than salmonella, contributing to the understanding of their pathogenesis. After acquiring the images, perform single-cell analysis by opening the acquisition. czi file in ImageJ.

Split the channels by clicking the image and selecting Color, followed by Split Channels. For cell segmentation, choose the Z plane image in the C1 acquisition title window. Select the C1 acquisition title window and click on the image followed by the duplicate.

Then, write the number of the chosen Z plane. Write C1Zplane in the title box, uncheck the Duplicate stack, and click OK to duplicate the selected Z plane image only. A window called C1Zplane showing the selected Z plane image will open.

Now enhance the image contrast of the obtained C1Zplane image by selecting Process and clicking Enhance Contrast. Adjust the saturated pixels, check Normalize, and click OK to apply the contrast enhancement technique to obtain a contrast-normalized C1Zplane image. Go to the Process and click Find Maxima to segment the epithelial cells in contrast-normalized C1Zplane image.

Then, to attribute only one maxima point to every epithelial cell, flag the Preview point selection and set the noise tolerance. To obtain the C1Zplane segmented image, a new binary mask-like image that shows each segmented particle per maxima point marked, flag the Exclude edge maxima, select the segmented particles as Output type, and click OK.Next, create a mask of segmented cells in the C1Zplane segmented image. Click Analyze, followed by Analyze Particles, then select the options Show masks and Exclude on edges.

Set the size interval and click OK to obtain the mask of C1Zplane segmented binary mask. Click the lookup table and select Invert lookup table in the toolbar. To threshold the contrast-normalized C1Zplane image, go to the image, select Adjust, select Threshold, and check the dark background.

Set the auto-threshold setting to default and choose the red option. Adjust the minimum cutoff value until cells appear completely red, leaving a dark background. Click Apply to convert the contrast-normalized C1Zplane image into a binary image with cells in white and the background in black.

Further, to correct the cell segmentation in the mask of C1Zplane segmented binary mask, select Process and click on the Image Calculator. Then, select the mask of C1Zplane segmented as image one, choose the operation AND in the dropdown list, and select the thresholded contrast-normalized C1Zplane as image two. Click OK and ImageJ will automatically name the output image as results of Mask of C1Zplane Segmented.

To label every single segmented cell of C1Zplane as a region of interest, click Analyze, followed by Analyze Particles. Adjust the size as demonstrated earlier and select the show Nothing option. Add all the particles to the ROI manager tool by checking Add to Manager.

Also, check the Exclude on edges and click OK.Save the ROI data as roi-cells-acquisitiontitle. zip through the ROI manager menu by clicking More and selecting Save. After finishing the cell segmentation, work on the C2 acquisition title window to analyze vacuolar salmonella.

First, go to the Process, click on the Image Calculator, select the C2 acquisition title as image one, and choose the operation Subtract in the dropdown list. Then, select the C3 acquisition title as image two, check Create new window, and click OK.Apply subtraction to the whole Z stack by clicking Yes in the Process Stack window. Duplicate the results of C2 acquisition title window by clicking on the Image menu and selecting Duplicate.

Then check the duplicate stack and press OK to obtain the results of C2 acquisition title one window. Apply Gaussian blur to the results of C2 acquisition title one window by going to the Process, clicking the Filter, and selecting the Gaussian blur. Leave the Sigma value as two or customize it.

Click on OK and apply Gaussian blur to the whole Z stack by clicking Yes in the Process stack window. Subtract the Gaussian filtered results of C2 acquisition title one to the results of C2 acquisition title by clicking on Process and selecting the Image Calculator. Now select the results of C2 acquisition title as image one, choose the operation Subtract in the dropdown list, select the results of C2 acquisition title one as image two, and click OK.Apply subtraction to the whole Z stack by clicking Yes in the Process Stack window to obtain a window automatically named by ImageJ as the results of results of C2 acquisition title.

To separate salmonellae from the background of the results of results of C2 acquisition title, click the image, select Adjust, and click the Threshold. Then, check the dark background and set the auto-threshold to Otsu. Adjust the minimum cutoff value until salmonellae appear completely red, leaving a dark background.

Click on Apply and a window named Convert to Binary will open. Check the black background and click on OK.Select the middle Z plane corresponding to the intracellular salmonellae focus plane in the results of results of C2 acquisition title binary window. Click the image followed by Stacks and then select Set Slice.

Now write the number of the Z plane of choice and click OK.Next, set the measurements to record for each ROI by clicking Analyze and selecting Set Measurement. Then, check the area corresponding to the total area of each ROI. To record the area fraction occupied by thresholded pixels for each ROI, check the Area Fraction and Limit to threshold.

Check on the Display label to label every single ROI with the image title, channel, Z plane, and X/Y coordinates. Use the chosen Z plane of intracellular salmonellae to record labels, area, and percentage area occupied by salmonellae for every single cell. Open the roi-cells-acquisitiontitle.

zip file by clicking the File menu and selecting Open. Then, click on Measure in the ROI Manager menu. The output is a table reporting the selected measurements.

Finally, save the results table by selecting the file and clicking Save As in the results window. The cellular invasion, the vacuolar load, and the cytosolic hyper-replication of salmonella were visualized inside epithelial cells. The area analysis revealed a significantly higher infection ratio in Salmonella typhimurium compared to both Salmonella derby strains, demonstrating the efficacy of area analysis in detecting differences in the ability of salmonella strains to colonize epithelial cells.

Salmonella derby sipA-deleted mutant strain displayed a reduced hyper-replication ratio compared to Salmonella derby wild-type, consistent with the role of sipA in inducing hyper-replication. No hyper-replication difference was observed between Salmonella typhimurium and Salmonella derby wild-type. The single-cell analysis showed no significant difference in the percentage of infected cells in Salmonella derby strains compared to Salmonella typhimurium.

Otherwise, Salmonella derby strains generated a mean vacuolar load significantly lower than Salmonella typhimurium. No difference was observed between Salmonella typhimurium and Salmonella derby wild-type, while Salmonella derby sipA displayed a significant decrease in hyper-replication, consistent with the result of the area analysis. Salmonella includes several serovars with diverse virulence.

Investigating large numbers of strains by this fast and automated method contributes significantly to understanding this pathogen's real virulence potential.