As you are aware, Acridine orange is an intercalating dye that can permeate both live and dead cells. AO will stain all nucleated cells to generate green fluorescence. Propidium iodide can only enter dead cells with poor membrane integrity, so it will stain all dead nucleated cells to generate red fluorescence.
So, it is basically straightforward to calculate the cell non-viability using the PI. However, the AO enters all cells regardless of alive or dead. So how can I analyze the raw data to get meaningful cell viability/non-viability percentages?
I have positive and negative controls, and blank wells as well.
Thank you in advance.
The principle would be that acridine orange staining represents the total cell population, whereas propidium iodide staining represents the dead cell population. The ratio of propidium iodide fluorescence to acridine orange fluorescence would represent the proportion of dead cells.
The problem is to calibrate the measurement. I believe this sort of measurement is normally done by either microscopy or flow cytometry, so that individual cells are examined or counted. Making the measurement in bulk using a fluorescence plate reader does not allow that, so some sort of calibration standards are needed (unless you are using an imaging plate reader).
Adam B Shapiro Thank you for your answer. Can I share with you the attached raw data and my data analysis when taking the PI fluorescent measurements, to confirm with you if these are correct? Your feedback is highly appreciated.
I need to present only the PI results and disregard the AO part. So, I'm expecting to have a decrease in cell viability between the negative control and treated sample. However, I did not get any difference. Given that a reduction in number of the cells was obvious under the microscope Moreover, we conducted Resazurin and MTT on the same, and we got a significant decrease in cell viability between 30-50%. However, the PI results are not compatible with this.
Many thanks for considering.
It looks like there is a slight difference of a few % between the untreated control and the two treated samples, but it is probably not statistically significant. This is probably because of the high background fluorescence from the PI in the untreated control. Did you replace the medium after the treatment with PI-free medium before making the measurement? PI may also be sticking to the plastic of the plate, resulting in a high background. You can test for this by treating wells that have no cells in them with PI as in the experiment.
Adam B Shapiro Many thanks for your prompt reply.
No, I did not discard the media after adding the PI for all wells in the plate. Neither, I added new media before taking the fluorescent readings.
So, on the day I do the experiment, you suggest adding PI. After 5-10 minutes of incubation to add new media. Then do the fluorescence reading?
Would you recommend doing PBS wash to the wells before adding the fresh media?
On this occasion to be clear, I added a large volume of the AOPI stain (20 Microliters). We are planning to use around 1-8 Microliters only of AOPI stain in the next run.
Once again, thank you for your insightful answers and your time. Really helpful!
Yes to both questions.
I will say, however, that I do not expect you to get a very large signal with this bulk method.
Adam B Shapiro Sorry to bother you again, but I've been thinking about the option you mentioned about the imaging plate reader, and the problem of the background effect you thankfully mentioned.
I have two questions please if you don't mind:
1. Reference to the attached images, is it ok to use the (Scan mode: matrix scan Scan matrix dimension: 3 × 3 Scan width [mm]: 5) to obtain the results of both AO & PI? Or not possible?
2. If I use a blank as (media + AO/PI stain), and subtract the fluorescent reading from what I get, will this remove the effect? Given that I will wash all wells before taking the measurements with PBS, and then add only new media, then I will take the fluorescent reading, as you kindly suggested. Will this solve the problem too? Or because the cells are bulk, it would still be a problem?
P.S: We have (BMG Labtech FLUOstar Omega plate reader) that can do the "well scan" option.
Many thanks for your valuable collaboration.
1. I've never used the well scanning feature, and I'm uncertain how you would use the results in a calculation. Perhaps, if the cells are attached to the plate in a patchy way, you could use it to find the best patch (brightest fluorescence, I suppose). However, it would be better to revise the cell plating method to get a continuous cell distribution throughout the well in order to get more consistent results. In that case, there would be no benefit to well scanning, since you would get about the same result from every position within the well.
2. I think the method you described, especially replacing the medium, will improve the result from the previous results you showed me by reducing the background. One thing to watch out for is if the background due to the fluorescent dyes sticking to the plastic of the wells is higher in empty wells than in wells containing cells, because the plastic is blocked by the cells. I can't think of any way to control for this. It may not matter if you use the ratio of the fluorescence of the two dyes as the readout.
You also should consider the length of time and the concentrations of the dyes as variables to be optimized in the experiment.
- Badges
- Science method