Fast Colony Forming Unit Counting in 96-Well Plate Format Applied to the Drosophila Microbiome

Counting colony forming units, or CFUs, is a standard technique in microbiology, but it’s slow and it requires lots of agar plates. Our technique provides a hundred-fold advantage over traditional CFU counting methods in both time agar plates. We’ve applied the methodology to measure the gut microbiome of small animal models, particularly fruit flies.

This protocol is a simple, fast, and automated way to quantify CFUs in gnotobiotics flies. This protocol provides a workflow with DIY-fabricated instrumentation to photograph the CFUs and custom software to automate counting them. This protocol could be used in any microbiology experiment where more efficient methods of plating and CFU counting could be useful.

Handling the flies efficiently and spotting the plates require manual dexterity, so this should be practiced. Using the automated counting software requires judgment about the threshold, so the receipts should be compared with manual counts to identify the best parameters for your experiment Demonstrating the procedure will be Ren Dodge, a researcher from my laboratory who developed the technique. To begin, pour 0.5-micron glass beads onto the bead measuring tray.

Spread the beads on the tray so that all the wells are full and leveled. Then, brush off excess beads into a conical tube. Now, place a semi-skirted PCR plate upside down on the measuring tray and align it with the wells by fitting it into the indentation on the measuring tray.

Then, quickly flip it over to transfer the beads. Remove excess beads from the PCR plate surface. Ensure that all the wells contain beads.

If required, use a weigh scoop to add beads to a single well. Then, cover the plate with aluminum foil and autoclave. Before using the plate, add 100 microliters of PBS to each well immediately prior to adding the flies using the 96-channel pipetter.

For surface sterilizing the anesthetized flies, transfer flies from the vial into a 1.5-milliliter microcentrifuge tube using a small funnel. Immediately spray one milliliter of 70%ethanol into the tube, close the tube, and mix by inversion for 10 seconds. Then, aspirate the ethanol with a P-1000 pipette while avoiding aspirating any flies.

After final PBS wash, close the tube and tap it hard on the bench a few times with the cap side down so the flies go into the cap. After opening the tube, dispense flies into the wells using forceps and place one fly in each well. Keep the flies anesthetized by keeping the plate on ice while loading.

Remove stray beads close to the wells, as these can break the foil seal and cause leakage. Make sure the dull side of the sealing foil is oriented down on the plate and the shiny side faces up towards the heat sealer. Press the heat sealer firmly for five seconds and burnish with the hand applicator to secure the foil.

Homogenize the flies for five minutes. Ensure that the flies are completely homogenized and the solution turns colored. To remove liquid from the sealing foil, spin down the bead plate for 30 seconds in a mini plate spinner at 350 G.Remove the foil while holding the plate to avoid splashing droplets of fly homogenate.

Place three rectangular MRS agar growth plates with their lids open in the biosafety cabinet to dry for at least 10 minutes. Prepare two dilution plates by adding 100 microliters of PBS to each well of a sterile 96-well plate using a 96-channel pipetter. Load a rack of P-20 tips onto the 96-channel pipetter.

Make a 1:10 dilution by aspirating 11.1 microliters of homogenate from the sample plate. Now, dispense it into the first dilution plate containing 100 microliters of sterile PBS per well. Keep the dilution plate on the plate shaker for 10 seconds at 600 RPM.

Mix again by pipetting up and down at least 10 times. Transfer 11.1 microliters from the first dilution plate to the second dilution plate and repeat the mixing steps for any further dilutions. For spotting plates, load two microliters from each well into the 96-channel pipetter.

Lower the pipetter head slowly onto the plate, avoiding stabbing into the agar, and ensure all the spots have been dispensed. Then, check that the liquid spots quickly soak into the agar and do not run together. Repeat the plating process for the remaining dilution plates as described in the manuscript.

Once the liquid has been completely absorbed into the agar, invert plates and incubate them until the colonies have reached optimal size. Organize the plates logically and keep them in a specific order while photographing. Orient them so A1 is in the upper left corner for all the plates.

After removing the plate lid, place the plate on the stage with A1 oriented in the correct corner. Image the plates. Transfer the images to a computer and rename them, including the experiment name, the media type, the dilution factor, and other pertinent details.

Organize the images to be processed for quantification into a folder. The file names that distinguish the plates become column titles for each set of counts in the output spreadsheet. Make subfolders named Cropped and Receipts in the folder.

These folders will receive the output files. Now launch the Croptacular plugin from ImageJ and click OK to start cropping. A dialogue box to choose the source folder will pop up.

Click OK.Select the source folder and click Open. Then, click OK on the dialogue box for choosing the destination directory. Select the destination folder and press Open.

If the image is already straight, simply press Space. Otherwise, straighten the image by drawing a line along an edge that should become horizontal. Redraw the line as many times as needed if the image does not appear straight.

Next, draw a boundary box for the counting area and adjust its size and proportion until all the spots are within their cells. Drag the cursor outside the boundary box to refresh the grid. When the grid looks good, press Space.

Since the plugin assumes all the photos to be aligned identically, ensure that the next image automatically rotates to the same angle as the first one. If this is accurate, press Space to continue. Otherwise, straighten the image as before.

The grid also recalls the same position as the previous image. Adjust if necessary and press Space. Launch Count-On-It and select Gridiron.

Set the threshold colony intensity based on an upper and a lower pixel intensity value. Make the threshold as stringent as possible, while still selecting all the colonies. Click OK on the Action Required dialogue box when the threshold is satisfactory, then click OK to continue.

On the first image, the option is given to proceed or return to the setup menu. To proceed with the batch process, click OK.Then, select a folder to keep the receipts and results table. Use the Receipts folder created earlier.

After the first image, the following images will default to the same threshold as the previous settings. Click OK to use these settings or adjust the settings. Review the count receipts that are produced by the software and manually correct any counting errors.

The ImageJ plugin is statistically accurate, but errors do occur. Proof the receipts to examine outliers and identify miscounts. The software saves the CFU data as a CSV file in the same folder as the receipts.

For flies colonized with Lactobacillus plantarum, there was a significant decrease in bacteria for flies that were either transferred daily to sterile food, transferred daily, kept on sterile food four hours before analysis, or transferred daily and kept on sterile agar four hours before analysis compared to flies that were never transferred to sterile food after inoculation. Similar results were observed in Acetobacter indonesiensis-fed flies, where transferred, post-transferred, and cleared flies showed a significant reduction in CFUs, but washing did not have a measurable effect, suggesting that the reduction in CFUs was due to the clearance of bacteria from the gut rather than the cuticle. The CFU count in spots with 2 to 25 colonies per spot showed no difference as compared to the traditional method.

Spots averaging 35 colonies showed lower CFUs than the control spread plates. This reduction might be due to colonies overlapping in dense spots. In plate photos, the colony intensity was 300%higher than the background.

The Lactobacillus plantarum mCherry-positive colonies displayed 1, 000%higher red intensity than mCherry-negative colonies. Acetobacter indonesiensis GFP-positive colonies showed 200%higher green intensity than GFP-negative colonies. The Count-On-It plugin for ImageJ automates CFU counting from the plate photos.

Colonies are accurately detected, with comparable counts to manual counting. Count receipts can be used to identify miscounts. Changing size thresholds and fluorescence wavelengths allows different colony morphologies to be counted separately.

It is essential to make sure the plate is well-sealed before bead beading. It is also important to design positive controls to calibrate your measurements. We also use this technique to study in vitro populations of bacteria in 96-well plates, which lets us study mixed communities of bacteria.

This technique allows high-throughput screening approaches to measure gut colonization as well as biological variation in colonization.