Hi Abdenour,

I agree that autofluorescence is typically not a problem in flow cytometry. However, there are certain cases when it is. For example, if the expression level of the stained antigen is low (10 thousand/cell), the signal intensity can be comparable to the background, i.e. autofluorescence. However, if you simply want to measure mean intensities in the channels corresponding to your fluorophores, cell-by-cell correction for autofluorescence is typically not required. Subtraction of the intensity of your unstained control (FMO control) is sufficient. On the other hand, if you would like to perform calculations with the intensities (e.g. FRET measurements; this is what I did) there will be a significant problem if the intensity of the background is comparable to your signal (see slide 36 and 37 of the linked tutorial). We have published a paper in which such a cell-by-cell correction for autofluorescence improved the quality of FRET measurements (see attached paper).

Regarding your question about why you need to include the spillover of your "BV510" signal to the other channels. If you specify an unused channel for your autofluorescence signal, you can imagine this as if you labeled your cells with another dye measured in your autofluorescence channel, e.g. if you use two dyes (Alexa488, Alexa647), with your autofluorescence channel you now have three dyes (two Alexas and a virtual dye corresponding to cellular autofluorescence). You can then define a compensation factor characterizing the spillover of your autofluorescence into your Alexa488 and Alexa647 channels. Using these spillover factors you can subtract autofluorescence in the Alexa488 and Alexa647 channels on a cell-by-cell basis.

I hope I could help you a bit.

Peter

http://cytometry.med.unideb.hu/sites/default/files/files/fret_tutorial_peternagy.pdf

Article Long wavelength fluorophores and cell-by-cell correction for...

I agree that autofluorescence is seldom an issue when doing FACS.

However, if for example you are staining lungs cells, you will notice that alveolar macrophages are highly auto-fluorescent. If you only care about T-cells, gating out the auto-fluorescent Macrophages will solve your problem.

but if you want to study these macrophages you may have to keep an open channel like FITC and use unstained macrophages as positive for FITC and beads or other cells (lymph node cells which are mostly T and B cells) as true negative controls. I hope this helps.

Hi Abdenour,

I agree that autofluorescence is typically not a problem in flow cytometry. However, there are certain cases when it is. For example, if the expression level of the stained antigen is low (10 thousand/cell), the signal intensity can be comparable to the background, i.e. autofluorescence. However, if you simply want to measure mean intensities in the channels corresponding to your fluorophores, cell-by-cell correction for autofluorescence is typically not required. Subtraction of the intensity of your unstained control (FMO control) is sufficient. On the other hand, if you would like to perform calculations with the intensities (e.g. FRET measurements; this is what I did) there will be a significant problem if the intensity of the background is comparable to your signal (see slide 36 and 37 of the linked tutorial). We have published a paper in which such a cell-by-cell correction for autofluorescence improved the quality of FRET measurements (see attached paper).

Regarding your question about why you need to include the spillover of your "BV510" signal to the other channels. If you specify an unused channel for your autofluorescence signal, you can imagine this as if you labeled your cells with another dye measured in your autofluorescence channel, e.g. if you use two dyes (Alexa488, Alexa647), with your autofluorescence channel you now have three dyes (two Alexas and a virtual dye corresponding to cellular autofluorescence). You can then define a compensation factor characterizing the spillover of your autofluorescence into your Alexa488 and Alexa647 channels. Using these spillover factors you can subtract autofluorescence in the Alexa488 and Alexa647 channels on a cell-by-cell basis.

I hope I could help you a bit.

Peter

http://cytometry.med.unideb.hu/sites/default/files/files/fret_tutorial_peternagy.pdf

Article Long wavelength fluorophores and cell-by-cell correction for...

thank you all for your answers, very helpfull!!

in spite of paying much attention on autofluoresce, use single colour tubes and better to use isotype controls for INTRACELLULAR markers/cytokine. You can even keep some tubes without intracellular marker but having all surface markers. This will help you to identify the positive signals of dim marker (intracellular markers) 

Interesting answers to autofluorescence, thanks!

Nearly all is said to this topic, but I would like to address it a little bit too. I like the answer "I agree that autofluorescence is typically not a problem in flow cytometry. However, there are certain cases when it is."... I think this total hit the mark. The problem is not the autofluorescence, but the problem is to know when the autofluorescence is important to be looked at or to be neglected.

E.g. very important paper:

http://jb.asm.org/content/195/18/4067.long

And like Eric wrote: I also like to use some channels to exclude cells, which I do not want to have in my analysis. Or after compensation setup, I add an unused channel (like Dapi or APC), which sometimes help to exclude unwanted data points.

Cheers