The mean fluorescent intensity for fluorescent protein fusions and concentration is linear within the dynamic range of the microscope images see Article Quantitative dynamic imaging of immune cell signalling using...

However, with immunofluorescence the characteristics of your antibody-antigen reaction determine the level of fluorescence and this may or may not be linear. It will depend on the type of antibody (poly or mono clonal) if you are using a 2 stage protocol etc. I got best results when calibrating Western blots with recombinant protein using monoclonal antibodies. So directly conjugated monoclonal antibodies should be more likely to give a linear signal.

It will not be easy to test if your antibody is reporting the concentration correctly but you could test this by expressing the same protein as a fluorescent fusion from a transiently transfected vector and staining with your antibody using a different coloured probe and see if the two signals have a linear relationship.

Thank you very much for your helpful answer, David G Spiller . Good to know that FP fluorescence is proportional to FP concentration. I really like the suggestion of expressing FP-tagged protein and labelling it with the corresponding antibody. We will give that a try. That said, I see two issues with this approach:

If you have any ideas how to overcome these issues, please let me know.

Paul

Usually with a CMV driven vector transiently transfected in to a cell line you will get a huge range of fusion protein expression levels (usually beyond the dynamic range of a single image with most imaging platforms). The expression level will depend on the fusion protein half life, the time after transfection, - or more accurately nuclear access- and the number of copies of the vector introduced into cells. Cells will only start to express protein from the vector after mitosis and the levels will then tend to increase untill the cells are fixed.

Even if your endogeneous protein is cell cycle regulated in abundance then you should still get a good linear relationship at the top end of the expression levels, the regression line will just not go through zero. The non-transfected cells in your population will give you a good idea of the dynamic range of the cell cycle regulated endogenous protein levels. Incidently, after a transient transfection (say after 24hs with fugene6 ) transfected cells usually appear in pairs with similar expression levels. The distance between the pairs of cells can give you a crude estimate of the time since mitosis.

Phospho specific antibodies are more of a challenge - you could try expressing a phospho-mimetic mutant vector to see if that works with your antibody, if it is specific for a particular phosphorylation site?

We once managed to get a really good polyclonal phospho specific antibody to a serine in I kappa B alpha that on western blots was much more sensitive (at least 10x ) than the antibody to I kappa B alpha itself, which just goes to show how varied antibody affinities are.

Thank you very much for sharing your insight, David G Spiller . It will definitely help us to design our experiments.