Hi,
I transfected HEK cells with fluorescent ABCG2 transporter (multidrug resistance transporter) , it has a GFP probe attached, and I am performing a live cell confocal microscopy experiment, where I would like to determine the kinetics of Mitoxantrone (chemotherapeutic, far red fluorescence) accumulation in the plasma membrane of the transfected GFP-positive cells. I am doing it by measuring far red fluorescence over time. For the information, mitoxantrone is the substrate of this transporter and ABCG2 is pumping it out of the cell, but when I apply an inhibitor to it, the accumulation rate increases, proving that the inhibitor works. That is how I am testing some of the compounds designed to be potential ABCG2 inhibitors.
Now that I hopefully explained the idea, I can get to the quantification part. I am confused about how many cells/fields of view to use to have a considerable number of sample for quantification. How many transfected cells should I take picture of? 50, 100? Is it for example 100 cells in 3 different experiments, or it is 100 cells per experiment, 3 times ?
As I have to take pictures in certain time point, I can only have one field of interest, where I usually have around 15 transfected cells. How can I collect 100 of them?
Or should I use regions of interest, few of them per one cell, around the membrane and measure that? I am really lost, so please those with similar experience, help!!!
Kind regards,
Marija
You can measure the total fluorescence in the whole FOV for each treatment over time and normalise to cell number (i.e the number of transfected cells within that particular FOV) and that would average out any differences between individual transfected cell and also the number of transfected cells in the whole image view (most confocal software and ImageJ can do this for image stacks). You'll probably need replicates though for each treatment. You could ROI specific cells or membranes as long as they are representative of the FOV/treatment and you measure the same number of cells/membranes in every treatment. I would expect that this would give more variation though between your treatments as multiple cells in the same FOV could show slight differences. I suppose it doesn't matter which method you use as long as it's consistent across your whole analysis.
Hi Marija,
People tend to do one of two things when determining sample size. You can either refer to a number of replicates that are „conventionally accepted” in your field for a particular experiment, or alternatively use statistical calculations to determine an ideal sample size based on certain assumptions/preliminary data. In the absence of standard protocols, you could go down this second route. For this you’ll need to have some idea of what an „average” set of results from your experiment look like. The number of replicates and fields of view per replicate will depend on:
- the variability between your cells/regions within an experiment
- the variability between independent experiments
- the effect size (ie how big of a difference) you want to detect.
I would recommend you have a brief read (or watch this YouTube video https://www.youtube.com/watch?v=clbg10IwzYA) about power / sample size calculations so you get the theory, and then try an online calculator such as the one found here: https://www.stat.ubc.ca/~rollin/stats/ssize/n2.html
For the first point, the number of fields of views, the main question is whether your 15 cells you currently image accurately represent the entire well. You would need multiple fields if for example you notice that within a single well there are regional differences. If Mitoxantrone intensity is fairly evenly distributed I would not worry too much. If you are concerned, I was wondering if you could extend your field of view. If you quantify average intensity in the whole cytoplasm, you could probably get away with using a lower mag objective to capture a larger field of view and higher number of cells.
Regarding the variability between experiments, if you have already performed it a few times you should have an idea what the standard deviation is and you could plug that into the sample size calculator.
- Badges
- Science method