To add my two pence I have to agree to the importance of independent controls. However, one would want to use non-radioactive ones if possible, perhaps BRDU. As pointed out by Alice, I also would expect quite a lot of differences between different applications. For one there is biology, so cell lines, cell volume, state of activation... but also 5uM is not giving you the same amount of labelling mainly dependent on suspending medium, dye solubility (more problematic with the dichloro-CFDA-SE), dye uptake and dye conversion (intra and extracellular), extracellular and intracellular binding and number of binding sites, incubation temperature and time... Similarly I do not think one can generalise all the PKH dyes as there are a number of differences. So it would be a miracle if you would get similar results in different systems. So just to copy a method does not necessarily work. It is interesting to see that in Bill's case with the red PKH dyes the fluorescence decreases slightly for the positive cells from day o to day 5 and than starts actually increasing not only for the negative population but also for the positive cells on day 9. Now if I would know for sure that the underlying scale is a log one (haven't we just talked about that recently?) than I would say that the increase in fluorescence for the positive labelled cells is far more substantial than for the negative labelled cells. Now where they got their fluorescence shift from (assuming the medium was changed a couple of times over the time course of the experiment) is an interesting question. It would also have been interesting to see the unlabelled control experiment for those datasets in that context. Regards Gerhard -----Original Message----- From: Alice.L.Givan@dartmouth.edu [SMTP:Alice.L.Givan@dartmouth.edu] Sent: Monday, November 26, 2001 2:40 PM To: Cytometry Mailing List Subject: Re: CFSE I need to continue the discussion and to respond to the very valid points made by Andrew Wells. Unlike Andrew, we do have some good data showing CFSE-inhibiton of cell division (as viewed by tritiated thymidine uptake) -- with some specific antigen activators (but not with ConA as a polyclonal stimulator). This occurred at 5 micromolar CFSE concentration. I have also seen some good data from other labs using CFSE at lower concentrations where inhibition did not occur even with specific antigens. So I believe that CFSE concentration is an important factor (as would be expected). In our experiments, the CFSE labelling was done before cell activation (although I do take Andrew's point that labeling activated cells that have already made their "important" proteins may be an important additional difficulty that needs to be considered when using CFSE). We have been using human cells. I have some feeling that human cells may differ from mouse cells . I agree completely that CFSE, in many (but not all) protocols, gives much tighter distributions than do the PKH dyes. And therefore one can often do precursor frequency and proliferation calculations on CFSE-labeled cells without needing to resort to ModFit-type deconvolution of the histograms (although deconvolution might help in the frequent cases where even the CFSE histogram distributions from the dividing cells are not completely distinct). This deconvolution is always required with PKH-labeling and has the problems (especially related to non-linear log amps) that Andrew described. So, with respect to tight labelling, CFSE is much better than the PKH dyes. Again, there may be some human/mouse differences here. The work showing lovely CFSE peaks during cell culture is often in mouse systems. I could be wrong on this --- and would like to hear from people using CFSE labelling of human T-cells. Unlike Andrew, we find that PKH-labelling is very stable --- and we routinely look at cells cultured over 8 days. So, I don't feel that the stability of PKH labeling is a problem in our protocols. And the intensity of PKH labelling is fine (especially since we cannot go very high with the CFSE staining, without worrying about functional inhibition). We usually have to back off on the brightness of our PKH labeling so that we don't have compensation problems. One of the main factors that Andrew did not mention is that CFSE is much cheaper than the PKH dyes. So, my general opinion (with which Andrew might agree!), is that you should use the CFSE dyes as long as you have confirmed (with, for example, tritiated thymidine uptake assays) that CFSE does not inhibit cell proliferation with your cells under your staining conditions and with your . So the message is that, with either dye, you need to be careful that the staining is not affecting the very function that you are studying. Alice Alice L. Givan Englert Cell Analysis Laboratory of the Norris Cotton Cancer Center Dartmouth Medical School Lebanon, New Hampshire NH 03756 tel 603-650-7661 fax 603-650-6130 givan@dartmouth.edu
This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:01:41 EST